Friday, May 29, 2009

The Mouse Genome is "Finished"

The first draft of the mouse genome was published by the Mouse Genome Sequencing Consortium back in 2002. At the time it was the only available non-human genome sequence. Since then several dozen other draft sequences have been published and many more are in progress. You can view a complete list at NCBI: Mammalian Genomes.

A finished version of the human genome sequence was published a few years ago and up until this month it was the only one listed as "complete." Now you can add the Mus musculus (mouse) genome to the list of complete publicly available genome sequences (Church et al. 2009).

When scientists say that a genome is complete or "finished" they don't really mean it. What they mean is that the effort has reached the point of diminishing returns. They are confident that they have found almost all of the genes and most of the important bits but they're well aware of the fact that some parts of the genome are missing.

This figure from the Church et al. paper illustrates the extent of a "finished" sequence. The green chromosomes represent the original draft sequence. Unsequenced regions are shown in black. As you can see, there were many gaps in the original sequence—176,000 to be exact.

The blue chromosomes represent the "finished" genome sequence. There are a lot fewer black regions and they are mostly confined to the centromeres/telomeres at the top ends of the chromsomes. As is the case with the human genome, these regions are mostly repetitive DNA that resists assembly into large blocks [The Human Genome Sequence Is not Complete].

The reason why the Y chromosome is missing is because it was a female genome that was sequenced.

Mice have 19 autosomes (non-sex chromosomes, see karyotype above). When you compare the mouse and human genomes you can see right away that the sequences of chromosomes aren't conserved. What is conserved are large blocks of sequence that may be found on one particular mouse chromosome but on a completely different human chromosome.

The mapping of these conserved synteny relationships reveals a great deal about the evolution of human and mouse chromosomes from a common ancestor. For example, the yellow block of sequence at the tip of human chromosome 1 (below) is found on mouse chromosome 4. The other parts of mouse chromosome 4 are found on human chromosome 6, 8, and 9.

What this means is that there are large blocks of genes that have been preserved since the time of the common ancestor. There are 334 chromosomal breakpoints that define the blocks of homologous sequence between human and mouse. The rearrangements took place in both lineages and the frequency of such rearrangements seems to be similar in most mammalian lineages.

The current "build" of the mouse genome has 20,210 protein-encoding genes. This is a substantial reduction from the 22,011 genes predicted in the initial draft sequence. As a general rule, the number of confirmed genes declines with each improvement in the sequence. This is mostly due to the joining together of gene fragments that were misidentified in the first draft. The authors note that 30% of the genes in the "finished" sequence were disrupted by errors and gaps in the first draft. Some new genes are added because of the addition of new sequence data but this doesn't compensate for the genes that are removed.

A total of 2,185 new genes were added. Most of them are duplicates of genes previously identified in the original draft sequence. In fact, the biggest change in the "finished" sequence is the identification of 126,000 Kb (126Mb) of duplicated sequences that were not detected in the first draft. This makes the mouse genome—with about 5% of segmentally duplicaed sequence—similar to the human genome. Initially there were hardly any duplicated regions in the mouse genome leading to speculation that duplications were much more common in primates.

Almost half of the duplicated regions exhibit different copy numbers in various strains of mice. Since the sequenced genome comes from a highly inbred line of laboratory mice (C57BL/6J), it is possible that the o0bserved copy number differs substantially from wild-type mice.

The human genome has 19,042 protein-encoding genes. Of these 15,187 (80%) have clear orthologs in the mouse genome. (Orthologs are homologous genes in the same location. They are related by descent from a common ancestor.) The orthologous genes represent 75% of the mouse genes. Most of the remaining genes are not novel genes but duplicates of the orthologs.

Surprisingly, there were only 12,845 orthologous genes in the first draft sequence. The difference is due to the identification of mistakes in the earlier data where sequence and assembly errors led to the misidentification of conserved genes. What this means is that a substantial number of papers comparing humans and mouse genomes will need to be re-evaluated. Here's how the authors put it ...
The shortcomings of the initial draft assembly are readily apparent now that a more-complete genome assembly is available. Undoubtedly these have led to incomplete or inaccurate understanding of some aspects of mouse biology. The availability of high quality genome sequence for the mouse will lead the way in dismissing some commonly held misconceptions and, more importantly, in revealing many previously hidden secrets of mouse biology.
The total length of protein-encoding exons in the mouse genome is 33,500 Kb (33.5 Mb). The revised genome size is 2,660,000 Kb (2.66 Gb). Thus, protein-encoding regions represent only 1.3% of the genome. This is similar to the value in the human genome (1.1% or 32.6 Mb out of 3.08 Gb).

There are many important non-coding sequences including centromeres, telomeres, origins of replication, scaffold attachment regions etc. All genes have substantial regulatory regions that aren't counted in the 1.3% of the genome that encodes protein. In addition, there are hundreds of tRNA genes, ribosomal RNA genes, and genes for essential small RNAs.

Nevertheless, a substantial proportion of the mouse genome (>90%) appears to be junk DNA with no known function. Most of it (~50%) consist of active and degenerate transposons similar to the LINES and SINES found in all other mammalian genomes.

[Photo Credit: Oak Ridge National Laboratory]

Church, D.M., Goodstadt, L., Hillier, L.W., Zody, M.C., Goldstein, S., et al. (2009) Lineage-Specific Biology Revealed by a Finished Genome Assembly of the Mouse. PLoS Biol 7(5): e1000112. [doi:10.1371/journal.pbio.1000112 ]


  1. Mouse has 19 autosomes, 20 chromosome pairs total.

  2. If I live long enough...I'll have the last laugh!Contact: Sooike Stoops
    VIB (the Flanders Institute for Biotechnology)
    Saved by junk DNA

    Leuven, Belgium - VIB researchers linked to K.U.Leuven and Harvard University show that stretches of DNA previously believed to be useless 'junk' DNA play a vital role in the evolution of our genome. They found that unstable pieces of junk DNA help tuning gene activity and enable organisms to quickly adapt to changes in their environments. The results will be published in the reputed scientific Journal Science.

    Junk DNA

    'Most people do not realize that all our genes only comprise about 3% of the total human genome. The rest is basically one large black box', says Kevin Verstrepen, heading the research team. 'Why do we have this DNA, what is it doing?'.

    Scientists used to believe that most of the DNA outside of genes, the so-called non-coding DNA, is useless trash that has sneaked into our genome and refuses to leave. One commonly known example of such 'junk DNA' are the so-called tandem repeats, short stretches of DNA that are repeated head-to-tail. 'At first sight, it may seem unlikely that this stutter-DNA has any biological function', says Marcelo Vinces, one of the lead authors on the paper. 'On the other hand, it seems hard to believe that nature would foster such a wasteful system'.

    Unstable repeats

    The international team of scientists found that stretches of tandem repeats influence the activity of neighboring genes. The repeats determine how tightly the local DNA is wrapped around specific proteins called 'nucleosomes', and this packaging structure dictates to what extent genes can be activated. Interestingly, tandem repeats are very unstable - the number of repeats changes frequently when the DNA is copied. These changes affect the local DNA packaging, which in turn alters gene activity. In this way, unstable junk DNA allows fast shifts in gene activity, which may allow organisms to tune the activity of genes to match changing environments -a vital principle for survival in the endless evolutionary race.

    Evolution in test tubes

    To further test their theory, the researchers conducted a complex experiment aimed at mimicking biological evolution, using yeast cells as Darwinian guinea pigs. Their results show that when a repeat is present near a gene, it is possible to select yeast mutants that show vastly increased activity of this gene. However, when the repeat region was removed, this fast evolution was impossible. 'If this was the real world' the researchers say 'only cells with the repeats would be able to swiftly adapt to changes, thereby beating their repeat-less counterparts in the game of evolution. Their junk DNA saved their lives'.

  3. Charlie Wagner says,

    Leuven, Belgium - VIB researchers linked to K.U.Leuven and Harvard University show that stretches of DNA previously believed to be useless 'junk' DNA play a vital role in the evolution of our genome.

    Hi Charlie!

    Thanks for the reference. Most of us know about it already but it doesn't hurt to let everyone know that another little bit (<0.5%) of non-coding DNA has a function.

    Of course it has nothing much to do with the argument that the vast majority of our genome is junk. Are you confused about that?

  4. Interesting summary, Larry. Thanks.

  5. Thanks for the nice summary! I was surprised to learn recently that nobody seemed to be interested in finishing the rat genome - let's see if this has any influence on that.

  6. This comment has been removed by the author.

  7. "Of course it has nothing much to do with the argument that the vast majority of our genome is junk. Are you confused about that?"Hi, Larry...

    Confused? No.

    Because I know you're wrong.

    Nature 458, 240-241 (12 March 2009) | doi:10.1038/458240a; Published online 11 March 2009

    Transcriptomics: Rethinking junk DNA

    Very few people believe the "junk DNA" hypothesis any more. It isn't supported by the facts.

    Like I said...if I live long enough...

  8. D.melanogaster and C. elegans were also sequenced by 2002.