Genomics Forum

Web discussion forum on Genomics

Wednesday, May 10, 2006

Genomics

Have a look at www.genomicsforum.co.uk for some general information about science.

Thursday, February 23, 2006

What's a Genome?

A genome is all of a living thing's genetic material. It is the entire set of hereditary instructions for building, running, and maintaining an organism, and passing life on to the next generation. The whole shebang.

In most living things, the genome is made of a chemical called DNA. The genome contains genes, which are packaged in chromosomes and affect specific characteristics of the organism.

Imagine these relationships as a set of Chinese boxes nested one inside the other. The largest box represents the genome. Inside it, a smaller box represents the chromosomes. Inside that is a box representing genes, and inside that, finally, is the smallest box, the DNA.

In short, the genome is divided into chromosomes, chromosomes contain genes, and genes are made of DNA.
The word " genome " was coined in about 1930, even though scientists didn't know then what the genome was made of. They only knew that the genome was important enough, whatever it was, to have a name.

Each one of earth's species has its own distinctive genome: the dog genome, the wheat genome, the genomes of the cow, cold virus, bok choy, Escherichia coli (a bacterium that lives in the human gut and in animal intestines), and so on.

So genomes belong to species, but they also belong to individuals. Every giraffe on the African savanna has a unique genome, as does every elephant, acacia tree, and ostrich. Unless you are an identical twin, your genome is different from that of every other person on earth—in fact, it is different from that of every other person who has ever lived.

Though unique, your genome is still recognizably a human genome. The difference is simply a matter of degree: The genome differences between two people are much smaller than the genome differences between people and our closest relatives, the chimpanzees.

Genomes are found in cells, the microscopic structures that make up all organisms. With a few exceptions, each of your body's trillions of cells contains a copy of your genome: the cells in your muscles, the cells in your brain, the cells in your blood, and so on.

Imagine all the trillions of genomes in your body, in other people's bodies, in cedar and apple trees, in walruses, forest mushrooms, and migrating birds: The whole world is full of genomes.

But if the genome is a commonplace thing, it is also quite powerful. A genome is information that affects every aspect of our behavior and physiology. Cooking dinner, digesting your food, talking, singing, sleeping—your genome has a hand in all these things.

A genome alone can't make a person, because we are also influenced by where we live, the human culture that surrounds us, and hundreds of other aspects of our environment. But the fact remains that you can't make a person without a genome.

Studying the human genome, therefore, is likely to give us insights into why some people die of heart disease and others die of cancer, why some people are comfortable schmoozing with a crowd of strangers and others are paralyzed by shyness, why some people have trouble keeping weight on while others have trouble keeping it off, and so on.

It's not difficult for scientists to get their hands on the human genome. They draw a bit of blood from people who volunteer to have their genomes studied, then use some simple laboratory procedures to break open the cells in the blood sample and extract the DNA. The body is constantly making more blood cells, so blood is a renewable genome resource.

Each human being contains a slightly different version of the human genome, but all human genomes are similar enough that we can learn about the human genome in general by studying the genomes of one or a few individual people.

Studying the genome can mean many different things. You can study a very small part of the genome or the genome as a whole. You can study the sequence of a gene, the function of a gene, the parts of the genome that regulate genes, or the DNA outside of genes. You can observe where genes are located in the genome, or investigate how different genes work together.

The next three chapters discuss several ways how scientists are studying the genome as a whole: sequencing genomes, mapping genomes, and studying the variation within genomes. Some of what we learn from this work might help doctors prevent and treat diseases better. Some of it might simply celebrate the variety that is the human species, or our unity with other forms of life. Some of it might open up possibilities we haven't even thought of yet.

A chromosome is a package containing a chunk of a genome—that is, it contains some of an organism's genes. The important word here is "package": chromosomes help a cell to keep a large amount of genetic information neat, organized, and compact.

Chromosomes are made of DNA and protein. Most living things have chromosomes that are linear, like bits of fat thread, and are kept in the nucleus, a sphere-shaped sac within the cell.

Source: http://www.genomenewsnetwork.org/resources/whats_a_genome/Chp1_1_1.shtml#genome1

Dark Biology

The world's most deadly killer is back. Although officially eradicated over 20 years ago, smallpox is back in the news and in our collective consciousness.

With war looming, biological attacks on the United States have been predicted. The smallpox vaccinations many of us had as children were effective for perhaps as little as five years. President Bush has ordered the inoculation of 500,000 military personnel; in January, he issued a call for ten million healthcare workers to be vaccinated voluntarily. But many of these professionals, fearful of the vaccine's side effects, are not participating, effectively stalling the immunization plan.

Readers of Richard Preston's latest book, The Demon in the Freezer, may find this alarming.

Preston's book is the third in his "Dark Biology" trilogy. It follows The Hot Zone, a non-fiction account of an Ebola virus outbreak, and The Cobra Event, a thriller about biological weapons and terrorism. In The Demon in the Freezer, Preston returns to non-fiction with an inside look at the history, devastation, and eradication of the smallpox virus. We also learn how smallpox has once again become a serious danger, the biggest bioterrorist threat we face.

The deadliest natural smallpox virus is known as Variola major. It is lethal and highly contagious in the air. One infection can cause twenty new cases, each of which can cause twenty more. Indeed, smallpox can start the biological equivalent of a runaway chain reaction.

"It has taken the world nearly 20 years to reach roughly 50 million cases of AIDS—variola could reach that point in 10 to 20 weeks," Preston notes.

He illustrates the problem with the story of a German man who became infected during a trip to Pakistan in 1970. When the man, infected with smallpox, appeared in an emergency room in Germany, seventeen cases of smallpox occurred in the hospital on the floors above—and only a handful of the victims had any direct contact with him. The German government ultimately vaccinated a hundred thousand people to stop the outbreak.

Smallpox kills about a third of the people it infects. The virus is thought to have killed more people than any other infectious disease, including the Black Death. Preston writes, "Roughly one billion people were killed in the last 100 years of the virus's activity on earth" (which ended in 1979).

The disease causes skin to puff up with blisters the size of hazelnuts, especially over the face. The book's physical descriptions of smallpox victims can be gruesome and are not for the faint of heart. In a tame passage from an early chapter entitled "Stripper", Preston relates, "the virus had stripped the skin off his body, both inside and out, and the pain would have seemed almost beyond the capacity of human nature to endure."
The book's most inspiring section recounts the World Health Organization's successful effort to wipe out variola from the planet. The eradication campaign ran in full force from 1965 to 1975. A scientist named D.A. Henderson and seven staffers supervised thousands of healthcare workers who labored tirelessly to contain outbreaks around the world. For example, in India, at the peak of the campaign, workers visited 120 million homes over the course of just a year and a half.

The eradication of smallpox is considered one of mankind's greatest achievements, and Preston tells this story well, using the same engaging, clear prose throughout the book.

The story of smallpox does not end with its eradication from nature in 1979. Officially, the virus now exists in only two places—a storage facility in Russia and the Centers for Disease Control in Atlanta. Preston presents convincing evidence, however, that several nations, including Iraq and Russia, have developed biological weapons using the smallpox virus.

Preston introduces us to Peter Jahrling, the top scientist at USAMRIID, the United States Army Medical Research Institute for Infectious Diseases, whose primary mission is to protect the public against a biological terrorist attack.
"If you believe smallpox is sitting only in two freezers," Jahrling tells Preston, "I have a bridge for you to buy. The genie is out of the lamp."

The book pits Jahrling and D.A. Henderson, who is now the chief bioterrorism advisor to the Secretary of Health and Human Services, against each other in a complex bioethical conflict. Henderson and his supporters want to destroy the last two "official" strains of the smallpox virus. Jahrling fights to keep the smallpox virus alive. He and his USAMRIID team reawaken the smallpox virus at the CDC, and conduct a series of controversial experiments by infecting monkeys with live Variola for new vaccine research.
Preston presents chilling evidence that scientists in Russia and perhaps other countries are working to genetically engineer a new smallpox virus that would be resistant to the existing vaccine. "A vaccine-resistant smallpox would be everyone's worst nightmare come true," the author writes. "We would be left trying to fight a genetically engineered virus with a vaccine invented in 1796."

Preston jumps around and follows a number of narrative strands throughout the book; at times the story feels disjointed and the large cast of scientific characters can be overwhelming. He begins and ends the book with interesting, detailed coverage of the investigation of the anthrax attacks of 2001. But he does not make strong connections between the current smallpox threat and those anthrax deaths, and ultimately these chapters feel like they belong in a different book altogether.

Nevertheless, Demon in the Freezer is gripping, painstakingly researched, timely, and, above all, frightening beyond words.

Toward the end, Preston recounts a secret meeting of experts on infectious diseases held in April 2002. Asked to create a model of the spread of smallpox, one expert made this prediction: "…if you don't catch the first guy with smallpox before he kisses his wife, it goes out of control. We could be dealing with hundreds of thousands of deaths. It will absolutely shut down international trade, and it will make 9/11 look like a cakewalk. Smallpox can bring the world to its knees."

Source: http://www.genomenewsnetwork.org/articles/03_03/demon_rev.php

Cloning the Clone of a Famed Japanese Bull

Researchers have successfully produced a genetic clone of a bull they cloned several years ago. The bull calf—a clone of a clone—was born in Japan more than four years ago, but the birth was kept a secret until now because the researchers wanted to be sure the calf was healthy.

Xiangzhong Yang of the University of Connecticut in Storrs and Chikara Kubota of the Kagoshima Prefectural Cattle Breeding Institute in Kagoshima, Japan, made the Guiness Book of World Records in 2000 when they reported that they had cloned one of the most famous bulls of all time: Kamitakafuku, a 17-year-old Japanese bull that has been used to inseminate more than 350,000 cows.

The cloning of Kamitakafuku in 1998 resulted in the births of six offspring, and four calves survived.

Now, the researchers have created another success for the record books. They have cloned Saburo, one of Kamitakafuku’s clones. The effort produced two live calves in 200l; one clone died shortly after birth due to anemia, but the second, called Sho-zaburo or “Saburo Junior,” has thrived for the past four years in Japan.

“We kept it a secret for four years,” says Yang. “It’s a controversial field, and we didn’t want to jump into a big debate if we produced an abnormal clone. We wanted to make sure he would grow normally and have normal fertility.”

Although clones of clones have been produced in mice, this is the first time it has been done in a large animal. Yang and his colleagues have attempted unsuccessfully to produce a third generation clone—the clone of a clone of a clone.

Yang is not sure why his cloning attempts have succeeded. He employs a technique he developed in which the nucleus containing all of the bull’s genetic information is gently squeezed out of the donor cell and then transferred to a waiting egg cell from which the nucleus had been removed.

In the past, most cloning experiments called for aspirating the nucleus, or sucking the nucleus out of the cell, a process that Yang believes causes damage to the chromosomes. Researchers in Korea who announced the cloning of the first human embryo earlier this year used Yang’s more gentle technique for removing the donor nucleus.

Kamitakafuku, the “father” of the first and second set of clones, died in 2001 at the age of 21. His first set of cloned offspring—Saburo, Taro, Jiro, and Shiro—grew to be adults. Saburo and Shiro have fathered offspring through artificial insemination, as well as the old-fashioned way.

Source: http://www.genomenewsnetwork.org/articles/2004/05/26/bullclone.php

Cloning the Clone of a Famed Japanese Bull

Researchers have successfully produced a genetic clone of a bull they cloned several years ago. The bull calf—a clone of a clone—was born in Japan more than four years ago, but the birth was kept a secret until now because the researchers wanted to be sure the calf was healthy.

Xiangzhong Yang of the University of Connecticut in Storrs and Chikara Kubota of the Kagoshima Prefectural Cattle Breeding Institute in Kagoshima, Japan, made the Guiness Book of World Records in 2000 when they reported that they had cloned one of the most famous bulls of all time: Kamitakafuku, a 17-year-old Japanese bull that has been used to inseminate more than 350,000 cows.

The cloning of Kamitakafuku in 1998 resulted in the births of six offspring, and four calves survived.

Now, the researchers have created another success for the record books. They have cloned Saburo, one of Kamitakafuku’s clones. The effort produced two live calves in 200l; one clone died shortly after birth due to anemia, but the second, called Sho-zaburo or “Saburo Junior,” has thrived for the past four years in Japan.

“We kept it a secret for four years,” says Yang. “It’s a controversial field, and we didn’t want to jump into a big debate if we produced an abnormal clone. We wanted to make sure he would grow normally and have normal fertility.”

Although clones of clones have been produced in mice, this is the first time it has been done in a large animal. Yang and his colleagues have attempted unsuccessfully to produce a third generation clone—the clone of a clone of a clone.

Yang is not sure why his cloning attempts have succeeded. He employs a technique he developed in which the nucleus containing all of the bull’s genetic information is gently squeezed out of the donor cell and then transferred to a waiting egg cell from which the nucleus had been removed.

In the past, most cloning experiments called for aspirating the nucleus, or sucking the nucleus out of the cell, a process that Yang believes causes damage to the chromosomes. Researchers in Korea who announced the cloning of the first human embryo earlier this year used Yang’s more gentle technique for removing the donor nucleus.

Kamitakafuku, the “father” of the first and second set of clones, died in 2001 at the age of 21. His first set of cloned offspring—Saburo, Taro, Jiro, and Shiro—grew to be adults. Saburo and Shiro have fathered offspring through artificial insemination, as well as the old-fashioned way.

Source: http://www.genomenewsnetwork.org/articles/2004/05/26/bullclone.php

Thursday, February 09, 2006

Genetics

You may have heard of Gregor Johann Mendel. He was the monk whose research on peas lead to the understanding of how organisms inherit traits from their parents. Mendel studied how certain physical traits were passed from one generation of peas to the next. This research lead to the modern scientific field of genetics.

Since his pioneering research, scientists have discovered how DNA, the genetic code that tells an organism how to work and grow, is copied from one generation and then rewritten and recombined for the next generation. If you major in genetics, you’ll look at inheritance (including hereditary diseases) and the genetic path of evolution.

Genetics is the study of how DNA is passed down from one generation to the next.

Monday, February 06, 2006

How does GM work?

For the genetic modification of crops, a plant that has the desired characteristic is first identified. The specific gene that produces this characteristic is located and cut out of the plant's DNA. A piece of bacterial DNA, called a plasmid, is joined to the gene to act as the carrier that gets the gene into the cells of the plant being modified.

Furthermore, a 'promoter' is also included with the combined gene and plasmid. This helps make sure the gene works properly when it is put into the plant being modified. This gene package is then inserted into a bacterium, which is allowed to reproduce to create many copies of the gene package.
Finally, the gene packages are transferred into the plant being modified. The plant tissue that has taken up the gene packages is then grown into full-size GM plants.

Thursday, February 02, 2006

Plant genomics and biotechnology

Genomics and biotechnology are the modern tools for understanding plants at the various biological and
environmental levels, as well as boosting classical plant breeding techniques. An array of novel technologies
have emerged that are now permitting researchers to identify the genetic underpinnings of crop improvement, namely the genes that contribute to the improved productivity and quality of modern crop varieties.

The much-debated genetic modification (GM) of plants is one of the biotechnologies used, depending on the
specific challenges to be addressed, but we should not make the fundamental mistake of equating agricultural
and plant biotechnology with GM alone. Genetic modification of plants is not the only technology in the toolbox of modern plant biotechnologies.

Application of these technologies will substantially improve plant breeding, farming and food processing. In
particular, the new technologies will enhance our ability to improve crops further and, not only make them more traceable, but also enable different varieties to exist side by side (known as crop co-existence) – enhancing the consumer’s freedom to choose between conventional, organic and GM food. Taking advantage of the genetic diversity of plants will not only give consumers a wider choice of food, but it will also expand the range of plantderived products, including novel forms of pharmaceuticals, biodegradable plastics, bio-energy, paper, and more.

Plant genomics and biotechnology could potentially transform agriculture into a more knowledge-based
business. This could strengthen the sector’s sustainability and profitability by reducing uncertainty and the
dependence of farmers on variable ecological and climatic conditions. Critics fear that modifying the genetic
make up of some plants could lead to the possible emergence of such complications as ‘superweeds’ or toxic
food. Although scientific investigation in new fields is never risk-free, with due scientific precaution and diligence, constant and careful monitoring, and strict regulation, adverse developments can be prevented.
A strong European science and industry base in this important sector will allow Europe to address a number of socio-economic challenges:

• Securing a healthy and safe food supply
Improving living standards, particularly in developed countries, and a growing world population are rapidly
boosting global demand for high quality and safe food. Food shortages and famines are currently localised
phenomena and can be addressed by improving the distribution of the world’s food output. However, as the
globe’s 6 billion inhabitants climb to more than 9 billion over the next half century, not only will this mean there are more mouths to feed but there will be less arable land with which to do it. This means that food distribution will have to become more equitable and farming will need to become both more productive and diversified. In addition, to respond to consumer expectations, the quality of plants has to be improved and their nutritional value boosted.

• Sustainable agriculture
We urgently need to make today’s chemical-intensive agriculture more sustainable while maintaining its
productivity. In fact, we need to increase yields and simultaneously reduce or optimise the amount of fuel,
fertilisers, pesticides and water used up in the process. The dual challenges of global climate changes and
increased seasonal weather instabilities are placing additional strains on the world’s agricultural capacity,
particularly as more marginal land is farmed.

• Green and pleasant land
Agricultural waste can be reduced to a minimum through the efficient use of bio-waste to produce biomaterials and bio-energy. As we run down our supplies of fossil fuels and their environmental impact grows, we will need to substitute them with renewable and environmentally friendly fuel sources. In addition, efficient land management will become increasingly necessary to ensure diversity of agricultural production, protection of the environment and conservation of natural resources and biodiversity.

• Competitiveness and consumer choice
A competitive global position for the EU in agriculture, biotechnology and food production will benefit
employment and economic growth across the Union. Developing new technologies and agricultural products can help the environment and have a positive impact on rural development. In addition, it would ensure a strong domestic and sustainable European food supply offering consumers a wide choice of healthy and diverse food.

Sowing the seeds of modern civilisation

Plants form the basis for almost all life on Earth, as living organisms in their own right or as food for animals. By
catching solar energy, unlike animals, they actually add to the Earth’s net store of energy. The unique process of
photosynthesis which converts atmospheric carbon dioxide, the main ‘greenhouse gas’, into carbohydrates via a
solar energy-driven process produces the biomass needed by most living organisms.
When people first started cultivating plants in the Middle East some ten thousand years ago, it profoundly altered
their way of life. With the birth of agriculture, humanity moved away from the hand-to-mouth existence of the
hunter-gatherer and was able to store food for leaner times. No longer needing to chase their food, people were
able to abandon nomadic life and live in settled communities, sowing the seed for modern civilisation.
Today, plants are used not only as food for humans and feed for cattle, but as industrial raw materials. They are
also valuable sources of vitamins, antioxidants, oils, fibres and carbohydrates. They provide the ingredients for
most drugs and are a major traditional energy resource. The petroleum that keeps modern society ticking is also
of plant origin, while modern biofuels promise to reduce the pollution caused by fossil fuels.
Rising populations and living standards are not only straining the world’s food production capacity, they have led
to the increasing use of non-renewable resources and synthetic materials, such as plastics (a petroleum
product). This has meant that more greenhouse gases – such as carbon dioxide – are being pumped into the
atmosphere, contributing to climate change. It has also meant that the world is producing waste faster than the
Earth can break it down. Environmental concerns and finite supplies of fossil fuels are making biomaterials –
which can be broken down easily by natural forces – and biofuels attractive alternatives.

Plant Genomics

Plant genomics holds the promise of describing the entire genetic repertoire of plants. Ultimately, plant genomics may lead to the genetic modification of plants for optimal performance in different biological, ecological, and cultural environments for the benefit of humans and the environment.

Now that the genome of the model plant Arabidopsis thaliana has been sequenced, plant scientists are faced with the daunting challenge of understanding what each of the ~25,000 genes of this model plant does and how these genes relate to those of other plant species, including crop plants.

The information derived from studies of plant genomics will help us understand how genes enable a plant to function, how the bewildering diversity of functions that distinguish plant species is related to simple changes in their genomes, and to what extent the structure, function, and evolution of plant genes show parallels in animals and microbes.

Cornell is an internationally recognized leader in the areas of genetic mapping, plant gene identification and quantitative trait analysis, plant evolution, molecular breeding, and plant genome database development, as well as plant molecular biology, pathology, and physiology. The Plant Genomics Focus Area - a partnership between the university, the New York State Agricultural Experiment Station at Geneva, and two institutes located on campus, the Boyce Thompson Institute and the USDA/ARS, Plant Soil and Nutrition Laboratory - aims to ensure Cornell's position as a major player in the genomics revolution that is sweeping the plant sciences.

In the current phase of implementation of its strategic plan, the plant genomics focus area has concentrated on three major areas:

plant biochemistry, with an emphasis on deciphering the synthesis, accumulation, and function of primary and secondary metabolites;

plant diversity, with the goal of evaluating and preserving natural biodiversity, and
plant genome informatics in coordination with the Computational Genomics Focus Area, with the goal of developing plant genome databases and specialized software for comparative genomics and for mining the voluminous data generated by genomics research.

Major activities:
The Institute for Genomic Diversity (director: Steve Kresovich) was established for the study of the molecular diversity of plants and its preservation for exploitation in agriculture and (as a source of pharmaceutical and therapeutic compounds) for the benefit of human health.
The USDA Plant Genome Informatics USDA-ARS Center for Comparative Genomics and Bioinformatics unit houses the single largest cluster of plant genome databases in the country (including GrainGenes, RiceGenes, SolGEnes, CabbagePatch, RoseDB, and RiceBlastDB), and aims to integrate databases and develop effective tools for sequence-based comparisons.

Facilities:
Renovation of Emerson Hall as a Plant Genomics Center that will house new faculty in plant genomics.

Accomplishments
Cornell now has the leading public research/training program in plant proteomics
Cornell received a $1.6 million NYSTAR grant to further strengthen plant proteomics
Cornell is now recognized as a world leader in comparative plant genomics and in certain aspects of developmental plant biology

Benefits of Agricultural Genomics

Indicators of Agriculture's Impact Technological and scientific advances account for the vast percentage of growth in the U.S. agricultural industry. Government economists have reported that scientific breakthroughs are responsible for at least 75 percent of the productivity growth since 1949. And it is the public that benefits from more abundant supplies and less costly production methods. One indication of this is that the U.S. consumer now spends only 16 percent of his or her disposable income on food, instead of the 46 percent that was required in the 1940s.

The rate of return on research and development investment has been proven. Even the most conservative estimates would indicate that the California economy could expect a $1.5 billion boost from the new jobs, income, and markets that this Institute would help to create.

There is no doubt that a "green revolution" is on the horizon. Understanding the structures and functions of the genomes of agriculturally important plants and animals, of major insect and nematode pests as well as the diseases they spread, and of the interactions between plant genes and the human body will enable researchers to breed or engineer superior commodities or novel products.

Examples of research occurring in UC labs today include:

Inserting genes in major agricultural pests so that the adults die off before they can cause damage to crops

Creating a clean-burning hydrogen fuel from algae

Isolating genes responsible for fruit tree dwarfing to enable producers to plant more trees per acre

Cloning a pathogen-resistant gene found in peppers and introducing it into tomato plants

Investigating cotton genes to identify which ones are responsible for creating fiber and which ones can be manipulated to create more fiber per boll

Creating cereals that are easier to digest and less likely to cause allergic reactions

Increasing the size of flowers and seeds produced by plants by inserting the ANT (Aintegumenta) gene, known to regulate plant organ size

Identifying the genes in vegetables and fruits that have anti-carcinogenic characteristics The potential of genomics research to boost the economy cannot be overstated. A conservative estimate is that publicly funded agricultural research at the University of California generates up to 20 percent of an annual return to the state. New uses and markets for crops, greater productivity, dramatic reductions in production costs due to less need for pesticides or less damage from pests, and more efficient use of land and water are just some of the expected outcomes. In addition, the Institute will share its discoveries, products, and technologies with the private sector and will train agricultural genomics specialists through its educational component.

Introduction
Comparative genomics is the study of the differences
and similarities in genome structure and organization
in different organisms. For example, how are the
differences between humans and other organisms
reflected in our genomes? How similar are the
number and types of proteins in humans, fruit flies,
worms, plants, yeasts, and bacteria? Essentially,
comparative genomics is no more than the application
of the bioinformatics methods described in Chapter
9 to the analysis of whole-genome sequences
with the objective of identifying biological principles,
i.e. biology in silico. In a sense this statement
greatly underplays the real value of comparative
genomics for, as the reader will soon see, it is
an extremely powerful technique and provides biological
insights that could not be achieved in any
other way.
There are two drivers for comparative genetics.
One is a desire to have a much more detailed understanding
of the process of evolution at the gross level
(the origin of the major classes of organism) and at a
local level (what makes related species unique). The
second driver is the need to translate DNA sequence
data into proteins of known function. The rationale
here is that DNA sequences encoding important
cellular functions are more likely to be conserved
between species than sequences encoding dispensable
functions or non-coding sequences. Until
recently it was thought that the ideal species for
comparison are those whose form, physiology, and
behavior are as similar as possible but whose genomes
have evolved sufficiently that non-functional
sequences have had time to diverge. More recently,
Bofelli et al. (2004) have shown that by comparing
genomes that are very distantly related, e.g. mammals
and fish, it is possible to identify conserved
sequences that, presumably, have a significant
function.
The formation of orthologs and paralogs are
key steps in gene evolution
In order to compare genome organization in different
organisms it is necessary to distinguish between
orthologs and paralogs. Orthologs are homologous
genes in different organisms that encode proteins
with the same function and which have evolved by
direct vertical descent. Paralogs are homologous
genes within an organism encoding proteins with
related but non-identical functions. Implicit in these
definitions is that orthologs evolve simply by the
gradual accumulation of mutations, whereas paralogs
arise by gene duplication followed by mutation
accumulation. Good examples of paralogs are the
protein superfamilies described in Chapter 16 (see
Fig. 16.6 and Table 16.1).
There are many biochemical activities that are
common to most or all living organisms, e.g. the
citric acid cycle, the generation of ATP, the synthesis
of nucleotides, DNA replication, etc. It might be
thought that in each case the key proteins would
be orthologs. Indeed, “universal protein families”
shared by all archae, eubacteria, and eukaryotes
have been described (Kyrpides et al. 1999). However,
there is increasing evidence that functional equivalence
of proteins requires neither sequence similarity
nor even common three-dimensional folds (Galperin
et al. 1998, Huynen et al. 1999). The existence of two
or more distinct sets of orthologs that are responsible
for the same function in different organisms is called
non-orthologous gene displacement. Now that close
to 200 different genomes have been sequenced it is
clear that gene displacement occurs within most
essential genes. That is, there are at least two biochemical
solutions to each cellular requirement.
Only about 60 genes have been identified where
gene displacement has not been observed (as yet)
and most of these encode components of the trancription
and translation systems (Koonin 2003).
CHAPTER 18
Comparative genomics
POGC18 11/9/05 2:30 PM Page 373
374 CHAPTER 18
Protein evolution occurs by exon shuffling
Analysis of protein sequences and three-dimensional
structures has revealed that many proteins are composed
of discrete domains. These so-called mosaic
proteins are particularly abundant in the metazoa.
The majority of mosaic proteins are extracellular or
constitute the extracellular parts of membranebound
proteins and thus they may have played an
important part in the evolution of multicellularity.
The individual domains of a mosaic protein are often
involved in specific functions which contribute to its
overall activity. These domains are evolutionarily
mobile which means that they have spread during
evolution and now occur in otherwise unrelated
proteins (Doolittle 1995). Mobile domains are characterized
by their ability to fold independently. This
is an essential characteristic because it prevents
misfolding when they are inserted into a new protein
environment. To date, over 60 mobile domains have
been identified.
A survey of the genes that encode mosaic proteins
reveals a strong correlation between domain
organization and intron–exon structure (Kolkman
& Stemmer 2001); i.e. each domain tends to be
encoded by one or a combination of exons and new
combinations of exons are created by recombination
within the intervening sequences. This process
yields rearranged genes with altered function and is
known as exon shuffling. Because the average intron
is much longer than the average exon and the recombination
frequency is proportional to DNA length,
the vast majority of crossovers occur in non-coding
sequences. The large number of transposable elements
and repetitive sequences in introns will
facilitate exon shuffling by promoting mismatching
and recombination of non-homologous genes. An
example of exon shuffling is described in Box 18.1.
The process of blood coagulation and
fibrinolysis involves a complex cascade of
enzymatic reactions in which inactive
zymogens are converted into active enzymes.
These zymogens belong to the family of serine
proteases and their activation is accompanied
by proteolysis of a limited number of peptide
bonds. Comparison of the amino acid
sequences of the hemostatic proteases with
those of archetypal serine proteases such
as trypsin shows that the former have large
N-terminal extensions (Fig. B18.1). These
extensions consist of a number of discrete
domains with functions such as substrate
recognition, binding of co-factors, etc.
and the different domains show a strong
correlation with the exon structure of the
encoding genes.
Box 18.1 Hemostatic proteins as an example of exon shuffling
0 100 200 300 400 700 800
Amino acid number
500 600
Plasminogen
u-PA
t-PA
Factor XII
Factor XI
Prekallikrein
Prothrombin
Protein C
Factor X
Factor IX
Factor VII
leader gla EGF kringle PAN fnl fnll protease
Fig. B18.1 Domain structures of
the regulatory proteases of blood
coagulation and fibrinolysis. The
different domains: gray, serine
protease domain; dark purple, EGFlike
domain; dotted, Gla domain;
cross-hatch, PAN domain; light
purple, fibronectin type II domain
(fn2); black, fibronectin type I
domain (fn1). (Adapted from
Kolkman & Stemmer 2001.)
POGC18 11/9/05 2:30 PM Page 374
Comparative genomics 375
Although mosaic proteins are most common in
the metazoa, they are found in unicellular organisms.
Because a large number of microbial genomes
have been sequenced, including representatives from
the three primary kingdoms (Archaea, Eubacteria,
and Eukarya), it is possible to determine the evolutionary
mobility of domains. With this in mind, Wolf
et al. (2000a) searched the genomes of 15 bacteria,
four archaea, and one eukaryote for genes encoding
proteins consisting of domains from the different
kingdoms. They found 37 examples of proteins
consisting of a “native” domain and a horizontally
acquired “alien” domain. In several instances the
genome contained the gene for the mosaic protein
as well as a sequence encoding a stand-alone version
of the alien domain, but more usually the standalone
counterpart was missing.
Comparative genomics of bacteria
By mid-2004 the website of the National Center
for Biotechnology Information listed 173 bacteria
(19 Archaea and 154 Eubacteria) whose genomes
had been sequenced (http://www.ncbi.nlm.nih.
gov/genomes/MICROBES/Complete.html). Simple
analysis of the sequence data reveals two features
of note. First, the genome sizes vary from 0.49 Mb
(Nanoarchaeum equitans) to 9.1 Mb (Bradyrhizobium
japonicum and two species of Streptomyces), i.e. a more
than 18-fold difference. Secondly, the gene density
is remarkably similar across all species and is about
1 gene per kilobase of DNA. This means that large
prokaryotic genomes contain many more genes than
smaller ones. By contrast, the human genome contains
only twice as many genes as Drosophila. So how
can we account for the size diversity of prokaryotes?
When the different genomes are arranged in size
order (Fig. 18.1) some interesting features emerge.
First, the archaebacteria exhibit a very much smaller
range of genome sizes. This could be an artifact of
the small number of genomes examined but more
probably reflects the fact that most of them occupy
a specialized environment and have little need for
metabolic diversity. The exception is Methanosarcina
acetivorans. This bacterium is known to thrive in a
broad range of environments and at 5.8 Mb has
the largest archaeal genome (Galagan et al. 2002).
Second, the smallest eubacterial genomes are found
in those organisms that normally are found associated
Caulobacter crescentus, Vibrio cholerae, Legionella pneumophila
Bacillus subtilis, B. halodurans
Mycobacterium tuberculosis
Escherichia coli K12, Shigella flexnerii
Agrobacterium tumefaciens
Yersinia pestis
Synechocystis sp., Marinobacillus iheyensis
Deinococcus radiodurans, Mycobacterium leprae, Corynebacterium glutamicum
Xylella fastidiosa
Staphylococcus aureus
Listeria monocytogenes
Listeria innocua
Escherichia coli 0157
Pseudomonas aeruginosa
Sinorhizobium meliloti
Mesorhizobium loti
Saccharopolyspora erythraea, Streptomyces coelicolor
Saccharomyces cerevisiae
Clostridium acetobutylicum
Salmonella typhi
Salmonella typhimurium
0 1 2 3 4 8
Genome size (Mb)
5 6 7 13
Eubacteria
9
Bordetella pertussis
Xanthomonas campestris
Pseudomonas putida
Streptomyces avermitilis
Archaeabacteria
Fig. 18.1 Genome sizes of archaebacteria, some eubacteria, and one prokaryote whose genomes have been completely
sequenced.
POGC18 11/9/05 2:30 PM Page 375
376 CHAPTER 18
with animals or humans, e.g. mycoplasmas, rickettsias,
chlamydiae, etc. Those organisms that can
occupy a greater number of niches have a larger
genome size. Not surprisingly, there is a good correlation
between genome size and metabolic and
functional diversity as demonstrated by the size of
the genomes of Bacillus and Streptomyces (formation
of spores, antibiotic synthesis), rhizobia (symbiotic
nitrogen fixation), and Pseudomonas (degradation of
a wide range of aromatic compounds).
The minimal gene set consistent with
independent existence can be determined
using comparative genomics
The genome of N. equitans is the smallest sequenced
to date (Waters et al. 2003) but this organism is an
obligate symbiont. This begs the question, what is
the minimal genome that is consistent with a freeliving
cellular organism? In reality, this is a nonsensical
question unless one specifies a defined set of
environmental conditions. Conceivably, the absolute
minimal set of genes will correspond to the most
favorable conditions possible in which all essential
nutrients are provided and there are no environmental
stress factors. If one ignores functionally important
RNA molecules and non-coding sequences, the
problem is one of defining the minimal protein set.
The first attempt at identifying the minimal protein
set was made by compiling a list of orthologous
proteins in Hemophilus influenzae and M. genitalium
(Mushegian & Koonin 1996). The expectation was
that this list would predominantly contain proteins
integral for cell survival as both bacteria are essentially
parasites and thus should have shed auxiliary
genes. Altogether 244 orthologs were identified but
this list is unlikely to be complete because of the
occurrence of non-orthologous gene displacements.
Some of these gene displacements can be inferred
because both organisms appear to have key metabolic
pathways that are incomplete. In this way,
Mushegian (1999) extended the minimal protein
set to 256 genes.
The problem with the above approach is that if
one is too strict in defining the degree of similarity
between two proteins required to constitute orthologs
then the minimal protein set is greatly underestimated.
A variation of the above method is to identify
orthologous groups, i.e. clusters of genes that include
orthologs and, additionally, those paralogs where
there has been selective gene loss following gene
duplication. When this approach was taken with
four eubacteria, one archaebacterium, and one yeast,
816 clusters of orthologous groups (COGs) were
identified. Of these, 327 contained representatives of
all three kingdoms (Mushegian 1999). Based on this
set of 327 proteins it was possible to reconstruct all
the key biosynthetic pathways. When the analysis
was repeated with sequence data from an additional
three archaebacteria and 12 eubacteria, the minimal
protein set was slightly reduced to 322 COGs.
Larger microbial genomes have more paralogs
than smaller genomes
Comparison of the P. aeruginosa (6.3 Mb) and E. coli
(4.5 Mb) genomes indicates that the large genome of
P. aeruginosa is the result of greater genetic complexity
rather than differences in genome organization.
Distributions of open-reading frame (ORF) sizes and
inter-ORF spacings are nearly identical in the two
genomes. If the larger genome of P. aeruginosa arose
by recent gene duplication one would expect it
to have a similar number of paralogous groups
compared to the other large bacterial genomes
and a larger number of ORFs in each group. In fact,
the number of ORFs in the paralogous groups in
Pseudomonas is similar to the other genomes. Thus
selection for environmental versatility (Box 18.2)
has favored genetic capability through the development
of numerous small paralogous gene families
whose members encode distinct functions. As a
general rule, one would expect that as the size of
the prokaryotic genome increases then the number
of paralogs also would increase, and this is what has
been observed (Table 18.1). Furthermore, the biochemical
bias in these paralogs reflects the biology of
the host organism (Box 18.2).
Analysis of all the prokaryotic genomes sequenced
to date has revealed two intriguing observations.
First, almost half the ORFs identified are of unknown
biological function. This suggests that a number of
novel biochemical pathways remain to be identified.
Secondly, approximately 25% of all ORFs identified
are unique and have no significant sequence similarity
to any other available protein sequence. Although
this might be an artifact of the small number of
bacterial species studied by whole-genome analysis,
it does support the observation of incredible biological
diversity between bacteria. More importantly,
it indicates that there are large numbers of new
protein families yet to be discovered, e.g. over 1000
proteins in each of Bacillus subtilis, E. coli, and
Deinococcus radiodurans!
POGC18 11/9/05 2:30 PM Page 376
Pseudomonas aeruginosa
Pseudomonas aeruginosa is a bacterium that is
extremely versatile both ecologically and
metabolically. It grows in a wide variety of
habitats including soil, water, plant surfaces,
biofilms, and both in and on animals including
humans. A major problem with P. aeruginosa
is its resistance to many disinfectants and
antibiotics. Pseudomonads are characterized
by a limited ability to grow on carbohydrates
but a remarkable ability to metabolize many
other compounds including an astonishing
variety of aromatics.
Analysis of the genome of P. aeruginosa
(Stover et al. 2000) reveals a general lack of
sugar transporters and an incomplete
glycolytic pathway, both of which explain the
poor ability to grow on sugars. By contrast, it
has large numbers of transporters for a wide
range of metabolites and a substantial number
of genes for metabolic pathways not found in
many other bacteria such as E. coli. As might
be expected for an organism with great
metabolic versatility, a high proportion of the
genes (>8%) are involved in gene regulation.
The organism also has the most complex
chemosensory system of all the complete
bacterial genomes with four loci that encode
probable chemotaxis signal-transduction
pathways. Finally, sequencing revealed the
presence of a large number of undescribed
drug efflux systems which probably account
for the inherent resistance of the organism to
many antibacterial substances.
Caulobacter crescentus
Caulobacter crescentus is a bacterium that is
found in oligotrophic (very low nutrient)
environments and is not capable of growing
in rich media. Not surprisingly, genome
sequencing (Nierman et al. 2001) has shown
that the bacterium possesses a large number
of genes for responding to environmental
substrates. For example, 2.5% of the
genome is devoted to motility, there are
two chemotaxis systems and over 16
chemoreceptors. It also has 65 members
of the family of outer membrane proteins that
catalyze energy-dependent transport across
the membrane. By contrast, the metabolically
versatile P. aeruginosa has 32 and other
bacteria fewer than 10.
The bacterium also has an obligatory life
cycle involving asymmetric cell division and
differentiation (Fig. B18.2). Thus it comes as
no surprise that genome sequencing reveals a
very high number of two-component signaltransduction
proteins, e.g. 34 histidine protein
kinase (HPK) genes, 44 response regulator
(RR) genes, and 27 hybrid (HPK/RR) genes.
In addition, the frequency of the GAnTC
target site for DNA methylation was much
less than would be expected if it occurred
at random.
Deinococcus radiodurans
This bacterium is remarkable for its ability to
survive extremely high doses of ionizing
radiation. For example, it can grow in the
presence of chronic radiation (6 kilorads/hour)
and withstand acute exposures to 1500
kilorads. The organism also is resistant to
dessication, oxidizing agents, and ultraviolet
radiation. These properties could be the result
of one or more of prevention, tolerance, and
repair. Genome sequencing (White et al.
1999, Makarova et al. 2001) has shown that
systems for the prevention and tolerance of
DNA damage are present but that the key
mechanism of resistance is an extremely
efficient DNA repair system. Although all
of the DNA repair genes identified in D.
radiodurans have functional homologs in
other prokaryotes, no other species has the
same high degree of gene redundancy. The
bacterium also has multiple genes for proteins
involved in exporting oxidation products of
nucleotides. Another important component
may be the presence of DNA repeat elements
scattered throughout the genome. These
repeats satisfy several expected requirements
for involvement in recombinational repair,
including that they are intergenic, they are
ubiquitous, and they occur at a frequency
that is comparable to the number of doublestranded
DNA breaks that can be tolerated.
Box 18.2 Correlation of genome sequence data with the biology of bacteria
Swarmer
cell
Stalked
mother cell
Fig. B18.2 The life cycle of Caulobacter crescentus.
POGC18 11/9/05 2:30 PM Page 377
378 CHAPTER 18
Because the DNA and protein sequence databases
are updated daily it pays to revisit them from time
to time to determine if homologs to previously
unidentified proteins have been found. It also pays
to re-examine sequence data as new and more
sophisticated bioinformatics tools are being developed.
The benefits of this can be seen from the work
of Robinson et al. (1994). They re-examined 18 Mb
of prokaryotic DNA sequence and uncovered more
than 450 genes that had escaped detection. A more
specific example is that of Dandekar et al. (2000)
who re-examined the sequence data for Mycoplasma
pneumoniae. They identified an additional 12 ORFs
and eliminated one identified previously and found
an additional three RNA genes. They also shortened
eight protein reading frames and extended 16
others.
Horizontal gene transfer may be a significant
evolutionary force but is not easy to detect
Horizontal, or lateral, gene transfer is the occurrence
of genetic exchange between different evolutionary
lineages. It is generally recognized that horizontal
gene transfer has occurred but there is considerable
debate about the extent of its occurrence. For
example, Gogarten et al. (2002) believe that it occurs
much more than has hitherto been recognized
whereas Kurland et al. (2003) feel that it has had
little influence on genome phylogeny. Now that so
many microbial genomes have been sequenced it
might be thought that detecting lateral gene transfer
would be easy but there are doubts about the validity
of some of the methods used to detect it. Basically,
two methods are used: the detection of sequences
with unusual nucleotide composition and the detection
of a gene, or genes, for a function that is totally
absent in all closely related species. For example,
analysis of the genomes of two bacterial thermophiles
indicated that 20–25% of their genes were
more similar to genes in archaeabacteria than those
of eubacteria (Aravind et al. 1998, Nelson et al.
1999). These archaeal-like genes occurred in clusters
in the genome and had a markedly different
nucleotide composition and could have arisen by
horizontal gene transfer.
Garcia-Vallve et al. (2000) have developed a statistical
procedure for predicting whether genes of a
complete genome have been acquired by horizontal
gene transfer. This procedure is based on analysis of
G + C content, codon usage, amino acid usage, and
gene position. When it was applied to 24 sequenced
genomes it suggested that 1.5–14.5% of genes had
been horizontally transferred and that most of
these genes were present in only one or two lineages.
However, Koski et al. (2001) have urged caution in
the use of codon bias and base composition to predict
horizontal gene transfer. They compared the ORFs
of E. coli and Salmonella typhi, two closely related
bacteria that are estimated to have diverged 100 million
years ago. They found that many E. coli genes of
normal composition have no counterpart in S. typhi.
Conversely, many genes in E. coli have an atypical
composition and not only are also found in S. typhi,
but are found at the same position in the genome, i.e.
they are positional orthologs.
Karlin (2001) has defined genes as “putative
aliens” if their codon usage difference from the
average gene exceeds a high threshold and codon
usage differences from ribosomal protein genes and
chaperone genes also are high. Using this method, in
preference to variations in G + C content, he noted
that stretches of DNA with anomalous codon usage
were frequently associated with pathogenicity
islands. These are large stretches of DNA (35–200 kb)
that encode several virulence factors and are present
in all pathogenic isolates of a species and usually
absent from non-pathogenic isolates. Of particular
relevance is that they encode an integrase,
are flanked by direct repeats, and insert into the
Genome size Percentage of proteins
Organism relative to E. coli belonging to paralogs
Pseudomonas aeruginosa 1.4 75
Escherichia coli 1 50
Caulobacter crescentus 0.88 48
Hemophilus influenzae 0.38 35
Mycoplasma genitalium 0.12 26
Table 18.1
Relationship between
paralogs and
genome size.
POGC18 11/9/05 2:30 PM Page 378
Comparative genomics 379
chromosome adjacent to tRNA genes (Hacker et al.
1997). In this respect, pathogenicity islands resemble
temperate phages and could have been acquired
by new hosts by transduction (Boyd et al. 2001).
Alternatively, spread could have been achieved by
conjugative transposons. There are many other
putative examples of horizontal gene transfer (see
Gogarten et al. 2002, for a list) but the evidence
that transmission occurred in this way is much
scantier than for pathogenicity islands with the possible
exception of RNA polymerase (Iyer et al. 2004).
The comparative genomics of closely related
bacteria gives useful insights into microbial
evolution
Now that so many microbial genomes have been
sequenced it is possible to undertake comparative
genomic studies between closely related bacteria
or distantly related bacteria. Both kinds of studies
are valuable because they reveal different kinds of
information. Studies on distantly related bacteria are
covered in the next section and here we cover only
studies on bacteria that are phylogenetically close.
The most detailed comparative analysis of related
bacteria has been undertaken on three genera of
the Enterobacteriaceae: Escherichia, Shigella, and
Salmonella (Chaudhuri et al. 2004). Initially a comparison
was made between one laboratory strain
of E. coli and two O157 enteropathogenic isolates
(Hayashi et al. 2001, Perna et al. 2001) and later this
was supplemented by inclusion of a uropathogenic
strain (Welch et al. 2002). These studies showed
that the genomic backbone is homologous but the
homology is punctuated by hundreds of lineagespecific
islands of introgressed DNA scattered
throughout the genome. Also, the pathogenic strains
are 590–800 kb larger than the laboratory strain
and this size difference is caused entirely by variations
in the amount of island DNA. Many of these
islands are at the same relative backbone position in
the different pathogens but the island sequences are
unrelated. A more surprising finding was that only
39% of the proteins that each strain encodes are common
to all of the strains. Furthermore, the pathogen
genomes are as different from each other as each
pathogen is from the benign strain. A later analysis
of the genome of Shigella flexneri, a major cause
of dysentery, indicated that this bacterium has the
same genome structure as E. coli and even should be
considered as a distinct strain of E. coli rather than as
belonging to a different genus (Wei et al. 2003).
As noted earlier, distinctive codon usage is considered
to be an indicator of horizontal gene transfer.
Analysis of the different E. coli genomes showed that
the islands had distinctly different codon usage and
a 3–4.5 fold higher use of certain rare codons. Of
the approximately 2000 genes that were found in
islands in the pathogens only about 10% of them
were shared. However, many of these shared genes
are related to genes associated with bacteriophages
or insertion sequences suggesting that they may have
been involved in horizontal gene transfer. Many of
the other, non-shared island genes encode known
pathogenicity determinants. When different uropathogenic
strains are compared, e.g. ones responsible
for cystitis, pyelonephritis, and urosepsis, many of
their island genes are unique to one strain too. These
results suggest that both pathogenic and nonpathogenic
strains of E. coli have evolved through a
complex process. The ancestral backbone genes that
define E. coli have undergone slow accumulation
of vertically acquired sequence changes but the
remainder of the genes may have been introduced by
numerous occurrences of horizontal gene transfer.
Salmonella species are considered to be close
relatives of E. coli and two serovars (S. typhi and
S. typhimurium) have been completely sequenced
(McClelland et al. 2001, Parkhill et al. 2001a)
and compared to the E. coli genome they share
(Fig. 18.2), with which they share extensive synteny.
As would be expected, the relationship between
S. typhi and S. typhimurium is very much closer
than between S. typhi and E. coli, although there still
are significant differences. There are 601 genes
(13.1%) that are unique to S. typhi compared with
S. typhimurium and 479 genes (10.9%) unique to
S. typhimurium relative to S. typhi. By contrast,
there are 1505 genes (32.7%) unique to S. typhi relative
to E. coli and 1220 genes (28.4%) unique to
E. coli relative to S. typhi. Another difference between
S. typhi and S. typhimurium is the presence of 204
pseudogenes in the former and only 39 in the latter.
In most cases these pseudogenes are relatively recent
because they are caused by a single frameshift or stop
codon. It is worth noting that complete sequencing
of closely related genomes facilitates the detection of
pseudogenes. This is because a frame-shift or premature
stop codon is only recognizable if the gene is colinear
with a functional homologous gene in another
genome. One biological difference between the
two Salmonella serovars is that S. typhi only infects
humans, whereas S. typhimurium can infect a wide
range of mammals. This may be related to differences
POGC18 11/9/05 2:30 PM Page 379
380 CHAPTER 18
in pseudogene content because many of the pseudogenes
in S. typhi are in housekeeping functions and
virulence components.
The bacterium Bacillus anthracis is of much current
interest as it is the causative agent of anthrax
and has been used as a bioterrorism agent. It has
long been considered to be closely related to B. cereus,
which can cause food poisoning, and B. thuringiensis,
which is pathogenic for certain insects. A comparative
genomic analysis of these three strains has
shown that while they differ in their chromosomal
backbone the major differences in pathogenicity are
due to plasmid-borne genes (Radnedge et al. 2003,
Rasko et al. 2004, Hoffmaster et al. 2004). Originally
it was thought that B. cereus lacked the plasmids
pXO1 and pXO2 that respectively encode the lethal
toxin complex and the poly-gamma-glutamic acid
capsule, both of which are key virulence factors.
However, similar plasmids have been found in nonpathogenic
B. cereus strains and only differ from the
corresponding ones from B. anthracis by the lack of a
pathogenicity island containing various toxin genes.
There have been a number of genomic comparisons
made between different species of Mycobacterium.
Of these, the most interesting is that between M.
tuberculosis and M. leprae, the causative organisms of
tuberculosis and leprosy (Table 18.2). Of the 1604
ORFs in M. leprae, 1439 had homologs in M. tuberculosis.
Most of the 1116 pseudogenes were translationally
inert but also had functional counterparts in
M. tuberculosis. Even so, there has still been a massive
gene decay in the leprosy bacillus. Genes that
Number of events
140
100
100
60
20
20
Indel size (number of CDS)
60
1
51+
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21–25
26–30
31–35
36–40
41–45
46–50
Present in S. typhi, absent from E. coli
Present in E. coli, absent from S. typhi
Present in S. typhi, absent from S. typhimurium
Present in S. typhimurium, absent from S. typhi
Fig. 18.2 Distribution of insertions and deletions in S. typhi relative to E. coli and S. typhimurium. The graph shows number of
insertion–deletion events plotted against the size of the inserted or deleted element (shown as number of genes), clearly indicating
that most of the events involve a small number of genes. Values above the lines represent genes present in S. typhi; values below the
line represent genes absent in S. typhi. Dark bars show the comparison with S. typhimurium; light bars with E. coli. (Redrawn with
permission from Parkhill et al. 2001b.)
Mycobacterium
Feature Mycobacterium leprae tuberculosis
Genome size 3,268,203 4,411,532
G + C (%) 57.79 65.61
Protein coding (%) 49.5 90.8
Protein coding genes (No.) 1604 3959
Pseudogenes (No.) 1116 6
Gene density (bp per gene) 2037 1114
Average gene length (bp) 1011 1012
Average unknown gene length (bp) 338 653
Table 18.2
Comparison of the
genomes of two
Mycobacterium spp.
(Reproduced from
Cole et al. 2001.)
POGC18 11/9/05 2:30 PM Page 380
Comparative genomics 381
have been lost include those for part of the oxidative
respiratory chain and most of the microaerophilic
and anaerobic ones plus numerous catabolic systems.
These losses probably account for the inability of
microbiologists to culture M. leprae outside of animals.
At the genome organization level, 65 segments
showed synteny but differ in their relative order
and distribution. These breaks in synteny generally
correspond to dispersed repeats, tRNA genes, or
gene-poor regions, and repeat sequences occur at
the junctions of discontinuity. These data suggest
that genome rearrangements are the result of
multiple recombination events between related
repetitive sequences.
Comparative analysis of phylogenetically
diverse bacteria enables common structural
themes to be uncovered
Certain structural themes start to emerge as more
and more bacterial genomes are sequenced and
comparisons made between these sequences. One
such theme is the presence of pathogenicity islands
in pathogens and their absence from non-pathogens.
Another is that chromosomal inversions in closely
related bacteria are most likely to occur around the
origin or terminus of replication (Eisen et al. 2000,
Suyama & Bork 2001). Finally, many genomes are
littered with prophages and prophage remnants but
the exact significance of these is not known.
The systematic comparison of gene order in bacterial
and archaeal genomes has shown that there is
very little conservation of gene order between phylogenetically
distant genomes. A corollary of this is
that whenever statistically significant conservation
of gene order is observed then it could be indicative of
organization of the genes into operons. Wolf et al.
(2001) undertook a comparison of gene order in all
the sequenced prokaryotic genomes and found a
number of potential operons. Most of these operons
encode proteins that physically interact, e.g. ribosomal
proteins and ABC-type transporter cassettes.
More important, this analysis enabled functions to
be assigned to genes based on predictions of operon
function (Chapter 23).
Comparative genomics can be used to analyze
physiological phenomena
The bacterium Deinococcus radiodurans is characterized
by its ability to survive extremely high doses of
ionizing radiation. Although the complete genome
has been sequenced this has not been sufficient to
provide a convincing explanation for the observed
physiological phenotype (see Box 18.2, p. 373). Part
of the problem is that there are no other organisms
that exhibit the same degree of radiation resistance
with which to make comparisons. However,
Makarova et al. (2003) have made more progress
with understanding the basis for hyperthermophily.
In this context, hyperthermophily is the ability to
grow at temperatures exceeding 75°C whereas thermophily
is the ability to grow in the range 55–75°C.
Complete genome sequences were available for 11
hyperthermophiles including eight archaea from six
distinct lineages and three bacteria from diverse
phyla. Sequences also were available for 14 thermophiles.
Initially a search was made for COGs
which met the following criteria:
1 The COGs must encode proteins and be found in
at least three hyperthermophiles.
2 The number of hyperthermophiles with a particular
COG should be greater than the number of
mesophiles.
3 More than 50% of the organisms with a particular
COG should be thermophiles.
Altogether, 290 COGs met the above search criteria
but most of them were found only in archaeal hyperthermophiles.
Therefore the search was refined so
that at least one eubacterial hyperthermophile had
to encode each COG. In this way 58 COGs were
identified as being associated with the hyperthermophilic
phenotype. These COGs encode a variety of
different cellular functions (Fig. 18.3) and include
previously uncharacterized protein families.
Comparative genomics of organelles
Mitochondrial genomes exhibit an amazing
structural diversity
Mitochondria are ubiquitous in eukaryotes and play
a key role in the generation of ATP through the
coupling of electron transport and oxidative phosphorylation.
Although the function of mitochondria
is highly conserved the structure of the mitochondrial
genome exhibits remarkable variation in conformation
and size (Fig. 18.4; see Burger et al. 2003 for
review). Whereas the mtDNAs of animals and fungi
are relatively small (15–20 kb), those of plants are
very large (200–2000 kb). Plant mitochondria
POGC18 11/9/05 2:30 PM Page 381
382 CHAPTER 18
rival the eukaryotic nucleus, and especially the plant
nucleus, in terms of the C-value paradox they present:
i.e. larger plant mitochondrial genomes do not
appear to contain more genes than smaller ones but
simply have more spacer DNA. Plant mitochondria
also have a large amount of DNA derived from the
chloroplast, the nucleus, viruses, and other unknown
sources. This process probably is facilitated by the
existence of an active, transmembrane potentialdependent
mechanism of DNA uptake (Koulintchenko
et al. 2003). The C-value paradox extends to plant–
animal comparisons, where the Arabidopsis mtDNA
is 20 times larger than human mtDNA but has less
than twice the number of genes (Fig. 18.5). Even
within a single genus, in this case different species
Known replication and
repair genes
Predicted DNA repair
genes
Translation related
genes
Transcription related
genes
Cellular processes
Metabolic functions
General function
Uncharacterized protein
Fig. 18.5 Mitochondrial genome size and coding content
across eukaryotes. Length of coding regions of authentic
mitochondrial genes (purple), introns, intronic ORFs, phagelike
reverse transcriptases, and DNA polymerases (blue), and
intergenic regions (green). Species names are: Reclinomonas
americana (jakobid flagellate); Rhodomonas salina (cryptophyte
alga); Marchantia polymorpha (liverwort, bryophyte); Cafeteria
roenbergensis (stramenopile flagellate); Arabidopsis thaliana
(flowering plant, angiosperm); Homo sapiens (vertebrate
animal); Metridium senile (cnidarian animal); Saccharomyces
cerevisiae (ascomycete fungus); and Plasmodium falciparum
(apicomplexan protist). Amoebidium parasiticum
(ichthyosporean protist); Jakoba libera ( jakobid flagellate);
and Chlamydomonas reinhardtii (green alga, chlorophyte).
Fig. 18.4 Size and gene content of mitochondrial genomes
compared with an α-Proteobacterial (Rickettsia) genome.
Circles and lines represent circular and linear genome
shapes, repectively. (Reprinted from Gray et al. 1999
by permission of the American Association for the
Advancement of Science.)
Plasmodium
Homo
C. reinhardtii
S. pombe
C. eugametos
Chondrus
Phytophthora
Ochromonas
Acanthamoeba
Tetrahymena
Prototheca
Allomyces
Reclinomonas
Jakoba
Marchantia
Arabidopsis
Rickettsia
Authentic mitochondrial genes
Introns, ORFs, plasmid-derived genes
Non-coding regions
Key:
Arabidopsis
Amoebidium
Marchantia
Jakoba
Saccharomyces
Reclinomonas
Rhodomonas
Cafeteria
Metridium
Homo
Chlamydomonas
Plasmodium
400
350
200
150
100
50
0
Mitochondrial genoms size (kbp)
Fig. 18.3 Functions of the 58
COGs associated with
hyperthermophily. (Figure
reproduced from Makarova et al.,
2003, Trends in Genetics 19,
172–6.)
POGC18 11/9/05 2:30 PM Page 382
Comparative genomics 383
of the yeast Schizosaccharomyces, there can be a
four-fold variation in the amount of non-coding DNA
(Bullerwell et al. 2003).
As a result of the steady accumulation of sequence
data it now is evident that mtDNAs come in two
basic types. These have been designated as “ancestral”
and “derived” (Gray et al. 1999) and their
characteristics are summarized in Table 18.3. It is
generally believed that mitochondria are the direct
descendants of a bacterial endosymbiont that
became established in a nucleus-containing cell
and an ancestral mitochondrial genome is one that
has retained clear vestiges of this eubacterial ancestry.
The prototypal ancestral mtDNA is that of
Reclinomonas americana, a heterotrophic flagellated
protozoon. The mtDNA of this organism contains 97
genes including all the protein-coding genes found
in all other sequenced mtDNAs. Derived mitochondrial
genomes are ones that depart radically from the
ancestral pattern. In animals and many protists this
is accompanied by a substantial reduction in overall
size and gene content. In plants, and particularly
angiosperms, there has been extensive gene loss but
size has increased as a result of frequent duplication
of DNA and the capture of sequences from the
chloroplast and nucleus (Marienfeld et al. 1999).
If mitochondria are derived from a bacterium,
what is the closest relative of that bacterium that
exists today? The current view is that it is Rickettsia
prowazekii, the causative agent of epidemic typhus.
This organism favors an intracellular lifestyle that
could have initiated the endosymbiotic evolution of
the mitochondrion. The genome of R. prowazekii has
been sequenced and the functional profile of its genes
shows similarities to mitochondria (Andersson et al.
1998). The structure, organization, and gene content
of the bacterium most resemble those of the
mtDNA of Reclinomonas americana.
Gene transfer has occurred between mtDNA
and nuclear DNA
The principal function of the mitochondrion is the
generation of ATP via oxidative phosphorylation. At
least 21 genes encode proteins critical for oxidative
phosphorylation and one would expect all of these
genes to be located in the mtDNA. Similarly, an
mtDNA location would be expected for the genes
encoding the 14 ribosomal proteins that are required
to translate mtRNA. However, sequence data indicate
that many mitochondrial genomes lack a number
of key genes (Fig. 18.6) and the missing genes
can be found in the nucleus. Functional transfer of
mitochondrial genes to the nucleus has stopped in
animals, hence their consistency in size. Part of the
reason for this is that further transfer is blocked by
changes in the mitochondrial genetic code. However,
this gene transfer continues to occur in plants and
protists because there is no genetic code barrier to
transfer. Note that it is not just intact genes that are
transferred for Woischnik & Moraes (2002) found
human mitochondrial pseudogenes in the nuclear
genome. Many of these pseudogenes comprised parts
of two adjacent mitochondrial genes.
In the case of the mitochondrial cox2 gene, transfer
to the nucleus is still on-going in the case of the
legumes (Palmer et al. 2000). Analysis of 25 different
legumes identified some genera in which the cox2
gene was located in the mitochondrion, some in
which it was nuclear, and some where it was present
Table 18.3 Properties of ancestral and derived mtDNAs. (Reprinted from Marienfeld et al. 1999 by permission of Elsevier
Science.)
Ancestral mtDNA
1 Many extra genes compared with animal mtDNA
2 rRNA genes that encode eubacteria-like 23S, 16S,
and 5S rRNAs
3 Complete, or almost complete, set of tRNA genes
4 Tight packing of genetic information with few
or no introns
5 Eubacterial-like gene clusters
6 Use standard genetic code
Derived mtDNA
1 Extensive gene loss
2 Marked divergence in rDNA and rRNA structure
3 Accelerated rate of sequence divergence in both
protein-coding and rRNA genes
4 Highly biased use of codons including, in some
cases, elimination of certain codons
5 Introduction of non-standard codon assignments
POGC18 11/9/05 2:30 PM Page 383
384 CHAPTER 18
in both genomes. In most cases where two copies of
the gene are present, only one gene is transcriptionally
active, although at least one genus was found in
which both genes are transcribed.
Adams et al. (2000a) studied the distribution of
the rps10 gene in 277 angiosperms and identified 26
cases where the gene has been lost from the mtDNA.
In 16 of these loss lineages, the nuclear gene was
characterized in detail. To be active in the nucleus, a
gene acquired from mtDNA must be inserted into
the nuclear genome in such a way that a mature
translatable mRNA can be produced. Moreover, the
resulting protein is made in the cytoplasm and must
be targeted to and imported into mitochondria. What
emerged was that in some cases pre-existing copies
of other nuclear genes have been parasitized with
the rps10 coding sequence. In several instances a
mitochondrial targeting sequence has been co-opted
to provide entry for the RPS10 protein back into
the mitochondrion but different nuclear genes provide
this sequence in different plants. In other cases,
the RPS10 protein is imported despite the absence
of an obvious targeting sequence. These results,
and similar findings for other mitochondrial genes
(Adams et al. 2001), provide confirmation that
nuclear transfer is on-going and has happened on
many separate occasions in the past. Nor is nuclear
transfer confined to mitochondrial genes for Millen
et al. (2001) have made similar observations with
chloroplast genes. Henze & Martin (2000) have
reviewed the mechanisms whereby this transfer
can occur.
Horizontal gene transfer has been detected in
mitochondrial genomes
In the previous section we discussed intracellular
horizontal evolution whereby genes moved between
the mitochondrion and the nucleus. However, crossspecies
acquisition of DNA by plant mitochondrial
genomes also has been detected. The first example
detected was that of a homing group I intron (Palmer
et al. 2000). These introns encode site-specific endonucleases
with relatively long target sites that catalyze
their efficient spread from intron-containing
alleles to intron-lacking alleles of the same gene in
genetic crosses. This intron has been detected in the
mitochondrial cox1 gene of 48 angiosperms out of
281 tested. Based on sequence data for the intron
and the host genome, it appears that this intron has
been independently acquired by cross-species horizontal
transfer to the host plants on many separate
occasions. What is not clear are the identities of the
donor and recipient in each individual case. By contrast
with this group I intron, the 23 other introns in
angiosperm mtDNA belong to group II and all are
transmitted in a strictly vertical manner.
More recently, Bergthorsson et al. (2003) have
reported widespread horizontal transfer of mitochondrial
genes between distantly related angiosperms,
including between monocotyledonous and
dicotyledenous plants. The genomic consequences
of these mtDNA-to-mtDNA transfers include gene
duplication, recapture of genes previously lost
through transfer to the nucleus, and a chimeric
(half-monocot, half-dicot) ribosomal protein gene.
Respiration and oxidative phosphorylation
Ribosomal RNAs
Transfer RNAs
Ribosomal proteins and EF-Tu
Protein import and maturation
RNA maturation
Transcription
Key:
Reclinomonas
Rhodomonas
Marchantia
Cafeteria
Cyanidioschyzon
Arabidopsis
Monosiga
Homo
Metridium
Podinomonas
Saccharomyces
Plasmodium
100
90
80
70
60
50
40
30
20
10
0
Number of genes
Fig. 18.6 Mitochondrial gene classes and their
representation across eukaryotes. Species names are:
Reclinomonas americana ( jakobid flagellate); Rhodomonas
salina (cryptophyte alga); Marchantia polymorpha (liverwort,
bryophyte); Cafeteria roenbergensis (stramenopile flagellate);
Cyanidioschyzon merolae (red alga); Arabidopsis thaliana
(flowering plant, angiosperm); Monosiga brevicollis
(choanozoan flagellate); Homo sapiens (vertebrate animal);
Metridium senile (cnidarian animal); Pedinomonas minor
(green alga, chlorophyte); Saccharomyces cerevisiae
(ascomycete fungus); and Plasmodium falciparum
(apicomplexan protist).
POGC18 11/9/05 2:30 PM Page 384
Comparative genomics 385
Comparative genomics of eukaryotes
The minimal eukaryotic genome is smaller
than many bacterial genomes
In determining the minimal genome we are seeking
to answer a number of different questions. What is
the minimal size of the genome of a free-living unicellular
eukaryote and how does it compare with the
minimal bacterial genome? That is, what are the
fundamental genetic differences between a eukaryotic
and a prokaryotic cell? Next, what additional
genetic information does it require for multicellular
coordination? In animals, what are the minimum
sizes for a vertebrate genome and a mammalian
genome? Finally, what is the minimum size of
genome for a flowering plant? Given that many
eukaryotic genomes contain large amounts of noncoding
DNA these questions have to be answered
by considering both genome size and the number of
proteins that are encoded.
The smallest eukaryotic genome that has been
sequenced is that of the obligate intracellular parasite
Encephalitozoon cuniculi (Katinka et al. 2001).
This has a genome size of only 2.9 Mb although its
close relative E. intestinalis may have a genome that
is even smaller (2.3 Mb). Genome compaction in
these organisms is achieved by a reduction in the
length of intergenic spacers and a shortness of most
putative proteins relative to their orthologs in other
eukaryotes. Even so, E. cuniculi has approximately
2000 ORFs, which is 7–8 times the number in the
minimal bacterial genome. The genome of the yeast
Schizosaccharomyces pombe has about 4800 ORFs
(Wood et al. 2002) but is unlikely to represent the
minimal free-living eukaryotic genome unless the
E. cuniculi genome has lost many more essential
genes than those metabolic and biosynthetic
ones already recognized. The multicellular fungus
Neurospora crassa has approximately 10,000 ORFs
(Galagan et al. 2003), about 25% fewer than the
fruitfly Drosophila melanogaster (Adams et al. 2000b).
Many of these genes do not have homologs in either
Saccharomyces cerevisiae or S. pombe (Borkovich et al.
2004) but exactly how many of them are essential
for multicellular existence remains to be seen.
Comparative genomics can be used to identify
genes and regulatory elements
As noted in Chapter 9 accurately identifying genes in
a complete genome sequence can be very difficult
and identifying regulatory elements can be even
harder still. A powerful method for finding functional
elements such as genes and regulatory regions
is to align orthologous genomic sequences from
different species and search out regions of sequence
conservation. The rationale for this approach is that
mutations in functional DNA will be deleterious and
thus counter-selected thereby resulting in a reduced
rate of evolution of functional elements. The two
most important factors affecting the results of a comparative
analysis are the amount of divergence being
captured and the phylogenetic scope of the aligned
sequences (Cooper & Sidow 2003). The amount of
divergence affects the power and resolution of the
analyses. The scope, which is defined as the narrowest
taxonomic group that encompasses all analyzed
sequences, affects the applicability of conclusions
and the generality of the results. For example, a
dipteran scope that includes Drosophila (fruitfly) and
Anopheles (mosquito) can be used to find elements
that were present in their common ancestor as
well as ones present before the diversification of
hexapods, arthropods, and metazoa (Fig. 18.7).
An example of a comparative analysis with a narrow
scope is the genomic comparison of S. cerevisiae
with three other species of Saccharomyces (Kellis et al.
2003). The gene analysis resulted in a major revision
of the S. cerevisiae gene catalog that affected
15% of all genes, reduced the total count by about
500 genes, and identified 43 new small ORFs (50–
99 amino acids). This latter finding is particularly
significant since small ORFs can only be considered
putative genes in the absence of function or conservation
in different species. A comparative analysis
with a more divergent scope is that between
the pufferfish (Fugu rubripes) and human genomes
(Aparicio et al. 2002). This identified almost 1000
putative genes that had not been identified in the
two published reports on the human genome
sequence.
The direct identification of regulatory elements is
very difficult since they are short (6–15 bp), tolerate
some degree of sequence variation, and follow few
known rules. Computational analysis of single
genome sequences has been used successfully to
identify regulatory elements such as promoters
associated with known sets of genes. However, this
approach is of relatively little value in identifying
other regulatory elements involved in gene expression
(enhancers, silencers) and chromatin organization
(insulators, matrix attachment regions). As the
examples below show, comparative analysis is much
POGC18 11/9/05 2:30 PM Page 385
386 CHAPTER 18
more useful in this respect. Comparisons within a
narrow scope are particularly useful as they permit
almost the entire genome to be scanned for regulatory
regions. In this way Kellis et al. (2003) were
able to recognize an additional nine regulatory protein
motifs in addition to the 42 that were already
known.
Enhancers are regulatory elements that upregulate
gene expression by sequence-specific positioning
of transcriptional activators. Enhancers can function
independently of position and orientation although
they generally are located within hundreds of
kilobases of their target genes. Using comparative
analysis, Spitz et al. (2003) discovered a cluster of
enhancer elements that are conserved between
mammals and Fugu. These enhancers coordinate
expression between Hoxd genes and nearby genes
that are evolutionarily unrelated.
Silencers are elements that are capable of
repressing transcription. Many are found near their
corresponding promoter but there are other types.
Sequencing of the chicken CD4 gene showed that it
is similar to the mammalian CD4 gene and has a
functional human silencer (Koskinen et al. 2002).
This level of distant conservation suggests that this
silencer has a fundamental role in controlling gene
expression.
Insulator elements are barriers that separate
domains within chromatin and confine the actions
of regulatory elements to their appropriate targets.
They can block the action of enhancers as well as
prevent the spread of chromatin condensation from
nearby regions. Farrell et al. (2002) discovered conserved
genomic regions that flank the β-globin loci
in mouse and man. These regions contain binding
sites for CTCF, a protein known to be important for
enhancer-blocking insulator activity.
Matrix attachment regions (MARs) are regions of
DNA that are involved in the binding to the nuclear
matrix. Glazko et al. (2003) aligned intergenic
sequences from mouse and man and identified conserved
segments. Further analysis showed that 11%
of these had sequence motifs characteristic of MARs
and that many of them precede the 5′ ends of genes.
This latter observation suggests a role in regulating
transcription.
Human
Mouse
Zebrafish
Fugu
Drosophila
Anopheles
Arthropod traits, e.g.
jointed appendages
Insect traits,
e.g. hexapodia
Dipteran traits,
e.g. halteres
Teleost traits
Tetrapod traits
Mammalian traits,
e.g. mammary glands
Vertebrate traits,
e.g. backbone, jaws
Metazoan traits,
e.g. multicellularity
Fig. 18.7 The importance of scope and the impact of shared ancestry on comparative sequence analysis. The tree describing the
relationships among six actively studied genomes is drawn in black (not to scale). Each colored line indicates the phylogenetic
scope that applies to each pair of species at the terminal nodes: red line, placental mammal scope; green line, teleost scope; blue
line, dipteran scope. Overlaps of the colored lines indicate shared ancestry and capture traits shared by the indicated scopes and,
by implication, shared functional elements. Open circles and associated text show various traits that exemplify the major animal
clades and the branch of the tree on which they arose.
POGC18 11/9/05 2:30 PM Page 386
Comparative genomics 387
Comparative genomics gives insight into the
evolution of key proteins
Koonin et al. (2004) have undertaken a comprehensive
evolutionary classification of the proteins encoded
in seven completely sequenced eukaryotic genomes:
three animals (man, nematode, and fruitfly), one plant
(Arabidopsis), a budding yeast, a fission yeast, and the
microsporidian E. cuniculi. In particular, they looked
for eukaryotic clusters of orthologous groups (KOGs)
and the results are shown in Fig. 18.8. The fraction
of proteins assigned to KOGs tends to decrease with
increasing genome size, except for the obligate parasite
E. cuniculi. By contrast, lineage-specific expansions
of paralogous groups show the opposite trend
with the largest numbers being in the higher eukaryotes.
Only a minority of KOGs have readily detectable
prokaryotic counterparts, indicating the extent of
innovation linked to the origin of eukaryotes.
A total of 131 KOGs were represented by a single
gene in each of the seven genomes. Since these KOGs
are present in the minimal genome of E. cuniculi they
must encode core biological functions. Nearly all
of them encode subunits of known multiprotein
complexes and many of them are involved in rRNA
processing, ribosome assembly, intron splicing, transcription,
and protein assembly and trafficking.
The evolution of species can be analyzed at
the genome level
The yeasts Saccharomyces paradoxus, S. mikatae, and
S. bayanus are estimated to have separated from
S. cerevisiae 5–20 million years ago. The genomes of
all four have been sequenced and Kellis et al. (2003)
have undertaken a comparative analysis. They found
a high level of “genomic churning” in the vicinity
of the telomeres and gene families in these regions
showed significant changes in number, order, and
orientation. Only a few rearrangements were seen
outside of the telomeric regions and these are
summarized in Table 18.4. All 20 inversions were
flanked by tRNA genes in opposite transcriptional
orientation and usually these were of the same isoacceptor
type. The role of tRNA genes in genomic
inversion has not been noted before. Of the nine
translocations, seven occurred between Ty elements
and two between highly similar pairs of ribosomal
genes.
At the gene level, five genes were unique to
S. paradoxus, eight genes unique to S. mikatae, and 19
unique to S. bayanus. Most of them encoded functions
involved in sugar metabolism or gene regulation.
The majority (86%) of these unique genes were
located near a telomere or a Ty element, locations
Number of proteins
14000
0
Number of species in KOGs
0 1 2 3 4 5 6 7
2000
4000
6000
8000
10000
12000
Hsa
Ecu
Sce
Spo
Ath
Dme
Cel
Fig. 18.8
Assignment of proteins
from each of the seven
analyzed eukaryotic
genomes to KOGs with
different numbers of
species and to LSEs.
0, Proteins without
detectable homologs
(singletons); 1, LSEs.
Species abbreviations:
Ath, Arabidopsis
thaliana; Cel,
Caenorhabditis elegans;
Dme, Drosophila
melanogaster; Ecu,
Encephalitozoon cuniculi;
Hsa, Homo sapiens;
Sce, Saccharomyces
cerevisisae; Spo,
Schizosaccharomyces
pombe.
POGC18 11/9/05 2:30 PM Page 387
388 CHAPTER 18
that are consistent with rapid genome evolution.
One gene was identified that appears to be evolving
very rapidly and across the four species showed 32%
nucleotide identity and 13% amino acid identity.
Functionally it appears to be involved in sporulation,
which in yeast is a stage in sexual reproduction. In
this regard, it is consistent with the observation that
many of the best-studied examples of positive selection
in other organisms are genes related to gamete
function. One gene also was identified that showed
perfect 100% conservation at the amino acid and
the nucleotide level. The latter observation is very
unusual given the redundancy of the genetic code
and suggests that the gene might encode an antisense
RNA.
Analysis of dipteran insect genomes permits
analysis of evolution in multicellular
organisms
The fruit fly Drosophila melanogaster and the malaria
mosquito Anopheles gambiae are both highly adapted,
successful dipteran species that diverged about 250
million years ago. They share a similar body plan
and a considerable number of other features but differ
in terms of ecology, morphology, and life style.
For example, Drosophila feeds on decaying fruit
while Anopheles feeds on the blood of specific hosts.
A number of obvious differences can be seen at the
whole-genome level (Table 18.5) but these give little
insight into the evolutionary process.
When the two genomes are compared at the
protein level (Zdobnov et al. 2002) five classes of
protein can be recognized (Fig. 18.9). A total of 6089
orthologs were identified in the two species and their
average sequence identity was 56%. By contrast there
is 61% sequence identity of orthologs between the
pufferfish and humans, which diverged 450 million
years ago. This indicates that insect proteins diverge
at a higher rate than vertebrate proteins. This could
be because insects have a much shorter life cycle and
may experience different selective pressures. When
the orthologs are classified according to gene ontology
it is not surprising to find that the proteins
involved in immunity show the greatest divergence
and structural proteins are the most conserved.
The “many-to-many” orthologs shown in Fig. 18.9
represent groups of genes in which gene duplication
has occurred in one or both species after divergence,
i.e. paralogy. These, and the homologs, probably
represent adaptations to environment and life strategies
leading to changes in cellular and phenotypic
features. For example, four Anopheles paralogs without
a counterpart in Drosophila are similar to the
human gene encoding leukotriene B4 12-hydroxy
dehydrogenase, an enzyme that can inactivate the
proinflammatory leukotriene B4. The anopheline
mosquito may have acquired this gene to facilitate
the taking of a blood meal. A total of 579 orthologs
were restricted to Anopheles and Drosophila and did
not even share domains with proteins identified
in the other organisms whose genomes have been
sequenced. Most of those that could be annotated
encoded specific odorant and taste receptors, cuticle
proteins, pheromone and pheromone-binding proteins,
and insect-specific defense molecules.
The dynamics of gene evolution can be analyzed
by comparing the intron and exon structure of
the 1:1 orthologs. For example, equivalent introns
Table 18.5 Genome statistics for the mosquito and
fruit fly.
Anopheles Drosophila
Genome size 278 Mb 165 Mb
Total exon length 10 Mb 13.6 Mb
Total intron length 22.6 Mb 12.9 Mb
Average introns per 3.5 4.7
gene
Average amino 548 649
acids per protein
Reciprocal Segmental
Species Translocations Inversions Duplications
S. paradoxus 0 4 3
S. mikatae 4 13 0
S. bayanus 5 3 0
Table 18.4
Genomic
rearrangements in
three yeast species
when compared with
S. cerevisiae.
POGC18 11/9/05 2:30 PM Page 388
Comparative genomics 389
in Drosophila have only half the length of those
in Anopheles whereas exon lengths and intron frequencies
are roughly similar. Approximately 55% of
Anopheles introns in 1:1 orthologs have equivalent
positions in Drosophila but almost 10,000 introns
have been lost or gained between the two species.
The rate of gain or loss of introns has been calculated
to be one per gene per 125 million years.
Given that the two diptera being studied are
estimated to have diverged 250 million years ago
one would expect that, in addition to changes in
exon/intron structure, there would be significant
variations in genome structure. Indeed, the gene
order of the 1:1 orthologs has only been retained
over very small distances and this is referred to
as microsynteny. However, at the macro level,
chromosomal arms exhibit significant remnants of
homology between the two species and major interarm
transfers and intra-arm shuffling of gene order
can be detected (Fig. 18.10).
18.6%
(a)
(c)
(d)
(b)
10.3%
15.9%
11.0%
Species specific, no significant hits to other species
Homologs, best hit in non-insect species
Homologs, best hit in insects
Many-to-many orthologs, incl. recent duplications
1:1 orthologs
Structural molecules (140)
Transporters (297)
Enzymes (1405)
Ligand binding or carrier (580)
Chaperones (40)
Nucleic acid binding (577)
Motor proteins (46)
Transcription regulators (281)
Enzyme regulators (70)
Signal transducers (303)
Defense and immunity (56)
UNKNOWN (3358)
14,000
12,000
10,000
8000
6000
4000
2000
0
Number of proteins
Sequence identities within
GO functional categories
11.1%
10.0%
17.9%
13.8%
44.2%
Drosophila
melanogaster
47.2%
Anopheles
gambiae
600
400
200
0
Orthologous pairs
0 200
Ag
Dm
400
Average protein size (aa)
600 800
0%
Sequence identity
20% 40% 60% 80% 100%
0%
Sequence identity
20% 40% 60% 80% 100%
Fig. 18.9 Analysis
of the proteome of
Anopheles gambiae
(Ag) and Drosophila
melanogaster (Dm).
(a) Classification of
the proteins according
to their conservation.
(b) Plot of the average
protein length for each
protein class in A.
(c) Histogram of the
sequence identities of
the 1:1 orthologs. (d)
Sequence conservation
of the 1:1 orthologs by
functional category.
POGC18 11/9/05 2:30 PM Page 389
390 CHAPTER 18
A number of mammalian genomes have been
sequenced and the data is facilitating analysis
of evolution
The genomes of humans, the mouse, and the rat
have been completely sequenced and good progress
is being made with the genome of the chimpanzee
(International Human Genome Sequencing
Consortium 2001, Venter et al. 2001, Mouse
Genome Sequencing Consortium 2002, Rat Genome
Sequencing Project Consortium 2004, The International
Chimpanzee Chromosome 22 Consortium
2004). Figure 18.11 shows an analysis of the three
completely sequenced genomes. About 1 billion
nucleotides (40% of rat genome) align in all three
species and this “ancestral core” contains 94–95%
of the known coding exons and regulatory regions,
which in turn represent 1–2% of the genome. A further
30% of the rat genome aligns only with the
2L
2R
3L
3R
2L
2R
X
X
3R
3L
Anopheles gambiae Drosophila melanogaster
Fig. 18.10 Homology of chromosomal arms in insects.
Each chromosomal arm is marked by a color shown around
its name (pairs of chromosomes with significant homology,
such as Dm2L/Ag3R, use the same color). Coloring inside
the schematic chromosome arms denotes microsynteny
matches to a region in the other species; the color shown is
the color of the chromosome containing the matching region
in the other species.
Rat
2500 Mb
Mouse
2400 Mb
209 298 173
38 358
825
115 88
378 319
386
162
681
81 116
471
Human 2800 Mb
33 43
23
24
13
2
30
3
22
Mouse
Ratspecific
Mousespecific
Human
Primatespecific
Simple
Rat
Ancestral to
human–mouse–rat
Ancestral to
mouse–rat
Genomic
DNA
Repetitive
DNA
Fig. 18.11 Aligning portions and origins of sequences in
rat, mouse, and human genomes. Each outlined ellipse is a
genome, and the overlapping areas indicate the amount of
sequence that aligns in all three species (rat, mouse, and
human) or in only two species. Non-overlapping regions
represent sequence that does not align. Types of repeats
classified by ancestry: those that predate the human–rodent
divergence (gray), those that arose on the rodent lineage
before the rat–mouse divergence (lavender), species-specific
(orange for rat, green for mouse, blue for human) and simple
(yellow), placed to illustrate the approximate amount of each
type in each alignment category. Uncolored areas are nonrepetitive
DNA – the bulk is assumed to be ancestral to the
human–rodent divergence. Numbers of nucleotides (in Mb)
are given for each sector (type of sequence and alignment
category).
POGC18 11/9/05 2:30 PM Page 390
Comparative genomics 391
mRNA
Reverse transcription
and insertion at new site
Promoter B
Promoter A
(e)
New splice sites evolve
Other TE sequences
degenerate
Mobile
element
(f)
Exon
(a)
Exons
Divergence
(b)
Organism A
Organism B
Transfer
Divergence
(c)
Fission Fusion
(d)
Duplication
Fig. 18.12 Mechanisms whereby new genes arise. (a) exon capture (exon shuffling); (b) duplication of a gene followed by
sequence divergence of the duplicate; (c) divergence of a gene following transfer to a new host; (d) fusion of two separate genes or
separation of two fused activities; (e) movement of a gene sequence via an mRNA intermediate followed by coupling to a promoter;
(f) capture of a transposable element (TE) followed by degeneration of the TE sequences. Real examples of these mechanisms can be
found in the review of Long et al. (2003).
POGC18 11/9/05 2:30 PM Page 391
392 CHAPTER 18
mouse genome and consists largely of rodent-specific
repeats. A further 15% of the rat genome comprises
rat-specific repeats. More genomic changes have
been detected in the rodent lineages than in the
human. These include approximately 250 large
rearrangements between a hypothetical rodent
ancestor and human, approximately 50 between
this ancestor and rat, and a similar number between
the ancestor and the mouse.
The rat, mouse, and human genomes encode similar
numbers of genes and the majority have persisted
without deletion or duplication since the last common
ancestor. About 90% of the genes have strict
orthologs in all three genomes but, compared with
humans, the rodents have expanded gene families
for functions associated with reproduction, immunity,
olfaction, and metabolism of xenobiotics. These
features are not surprising given what we know about
rodent biology! Almost all the human genes known
to be associated with disease have orthologs in the
rat and mouse genomes but there is one surprising
finding. Many SNPs causing disease in man are found
in mice but these mice are phenotypically normal.
The comparison of the human genome with that
of the chimpanzee is perhaps the most interesting of
all the genomic comparisons that can be made, as
the chimpanzee is our closest living relative. In particular,
comparative analysis should help to uncover
the genetic basis of cognitive function, bipedalism,
and speech development. At the time of writing the
complete chimpanzee genome was not available but
the 33.3 Mb sequence of chromosome 22 had been
completed (The International Chimpanzee Chromosome
22 Consortium 2004). Nearly 1.5% of the
chimpanzee genome had single base substitutions
when compared with its human equivalent (chromosome
21) in addition to approximately 68,000
insertions or deletions. These differences are sufficient
to generate changes in most of the 231 coding
sequences. In addition, different expansion of particular
subfamilies of retrotransposons was observed
between the different lineages, suggesting different
impacts of retrotransposition on human and chimpanzee
evolution. The full impact of these changes
remains to be deciphered.
Comparative genomics can be used to
uncover the molecular mechanisms that
generate new gene structures
The comparative analyses described in the previous
sections indicate that there is a general process of
new gene origination. This raises the question of the
origin of these new genes. Several molecular mechanisms
are known to be involved in the creation of
new gene structures (Fig. 18.12) and can operate
singly or in combination (Long et al. 2003). A good
example is jingwei, the first identified gene that
has originated recently (2 My) in the evolutionary
timescale (Fig. 18.13). This gene arose in the common
ancestor of two Drosophila species. The starting
point was the yellow emperor gene that duplicated to
give the yellow emperor and yande genes. Whereas
yellow emperor maintained its original functions,
yande underwent modification. In particular, mRNA
of the alcohol dehydrogenase gene retroposed into
the third intron of yande as a fused exon and recombined
with the first three yande exons. This formed
jingwei, a gene that is translated into a chimeric
protein.
Once created, new genes such as jingwei may
become modified beyond recognition. Examples of
this kind of change include domains involved in
protein–protein interactions such as von Willebrand
A, fibronectin type III, immunoglobulin, and SH3
modules (Ponting et al. 2000). These domains show
extensive proliferation in higher eukaryotes but
have only a distant relationship to homologs in
prokaryotes and lower eukaryotes.
Recombination of exons in 5’ part of yande and Adh plus
degeneration of 3’ part of yande to create jingwei
Duplication of yellow emperor gene
yellow emperor and yande genes
Retroposition of alcohol dehydrogenase gene into yande
Fig. 18.13 Genomic events leading to the formation of the
new gene jingwei.
POGC18 11/9/05 2:30 PM Page 392
Comparative genomics 393
Suggested reading
Kellis M., Patterson N., Endrizzi M., et al. (2003)
Sequencing and comparison of yeast species to identify
genes and regulatory elements. Nature 423,
241–54.
This is rapidly becoming a classic paper on the use of comparative
genomics to decipher genome sequences but it also
provides insights to the genomic changes that exist between
species.
Koonin E.V. (2003) Comparative genomics, minimal
gene-sets and the last universal common ancestor.
Nature Reviews Microbiology 1, 127–36.
Koonin E.V., Federova N.D., Jackson J.D., et al. (2003)
A comprehensive evolutionary classification of
proteins encoded in complete eukaryotic genomes.
Genome Biology 5, R7.
Eugene Koonin probably knows more than anyone about
extracting evolutionary information from sequence databases.
The two papers cited above are but a tiny sample of his
analyses.
Koonin E.V. (2005) Virology: Gulliver among the
Lilliputians. Current Biology 15, R167–9.
An analysis of the genome of a virus that is much bigger
than many parasitic bacteria.
Long M., Betran E., Thornton K. & Wang W. (2003)
The origin of new genes: glimpses from the young
and old. Nature Reviews Genetics 4, 865–75.
This is one of the few reviews that attempt to discuss where
new genes come from.
Paterson A.H., Bowers J.E., Chapman B.A., et al. (2004)
Comparative genome analysis of monocots and dicots,
towards characterization of angiosperm diversity.
Current Opinion in Biotechnology 15, 120–5.
Pedulla M.L., Ford M.E., Houtz J.M., et al. (2003) Origins
of highly mosaic mycobacteriophage genomes. Cell
113, 171–82.
These two papers cover topics not discussed in this chapter,
the comparative genomics of plants and viruses respectively,
and are well worth reading.
Each year, the January 1 issue of Nucleic Acids Research
is devoted to short reviews of the different molecular
biology and genomics databases. A considerable
number of these databases are for the purposes of
comparative genomics and all are linked to relevant
websites. An example is given below.
Useful website
http://colibase.bham.ac.uk
This is the website for coliBASE, an online database for
the comparative genomics of E. coli and its close relatives.
Now that a number of different strains of E. coli
have been completely sequenced it is clear that there is
much more genomic heterogeneity than expected.
POGC18 11/9/05 2:30 PM Page 393