Steven Salzberg (who is nominated for the Franklin award at bioinformatics.org) has an opinion piece in Genome Biology proposing wiki technology to help solve the problem of genome annotations getting out of date.
Continue reading Wikis for genome (re)annotation
A paper* this week from the Huffnagle lab argues that even though the human pathogenic fungus Cryptococcus neoformans can produce an oxylipin similar to prostaglandin, the authors were unable to identify any homologous cyclooxygenase genes in the genome. They showed through LC-MS-MS on supernatants from C. neoformans cells grown on arachidonic acid that molecules with activity similar to prostaglandin E2 are synthesized. BLAST searches of the genome could not identify any similar genes to cyclooxygenase genes including the PPo genes from Aspergillus which contain catalytic domains similar to mammalian cyclooxygenases.
So did C. neoformans evolve a new way to synthesize this enzyme which may act as a hormone and affect the host’s immune system? My cursory searches against other basidiomycete genomes did turn up homologs to these PPo genes in Ustilago and Coprinus so perhaps the enyzmes in the pathway have changed in the Cryptococcus lineage. Perhaps searches with protein structure of cyclooxygenases could pick up functionaly similar genes which would serve as good candidates which have little sequence similarity to the cannonical protein determined in humans.
* Paid access required for 6 months.
Here is an image of Neurospora crassa I took today in my first attempt at squashes. These are from strains that Dave Jacobson grew up with his constructs so I can’t take any credit other than playing with the microscope next door. Now my first attempt came out badly, so this is actually Dave’s prep as well. And these got dry so they aren’t as nices as they could be. For much nicer images, see N.B. Raju’s.
All that said, I hope these quick images give a hint at the extremely cool structures these fungi produce. These 8-chain ascospores are the result of meoisis that took place inside the perithecia (which was squeezed gently to release the rosettes [or not too gently in my case]).
( I was previous confused about the sample and had labeled this N. tetrasperma which has 4-chained ascospores [tetra] while this sample is crassa which has 8).
An NPR story on former Taylor Lab postdoc and current Harvard professor Anne Pringle airs tonight. They followed her, Ben, and Frank around collecting Amanita phalloides in Point Reyes in December. Poor Anne’s voice is going as she had a cold, but as usual she does a great job expressing her unbridled passion for mycology and biology.
The NPR newscast right after the report also has two briefs on medicinal research with fungi.
Ever wonder what goes on in a cow’s multi-chambered stomach? Probably not. I did think about it a little more after a trip to a teaching farm during grad school where we saw a cow with a fistula. This hole provides access to the cows stomach so that samples can be drawn of the community living in the gut and understand how the bovine stomach can digest the recalcitrant cellulose of grasses.
Of course all kinds of lovely things live in the dark, anaerobic environment. In fact there is a delicately balanced community of species. When cows are fed corn instead of grass this affects the rumen acid content and allows pathogenic E. coli like O:157 to survive. So far I don’t seen any JGI proposal for sequencing of the gut communities of rumens, but maybe that should be proposed.
Rumen fungi are probably not on your keyword list, but these fungi are extremomophiles living in highly anaerobic environment. A paper in Microbiology details an analysis of the genome of the anaerobic fungus Orpinomyces.
Splicing of pre-messenger RNA is necessary to remove introns and create well formed and translateable mRNA, but the purpose of introns still remains a mystery. One idea is they provide a role in the error checking machinery, or Nonsense Mediated Decay (NMD), by providing way-points during translation. A protein is deposited at the exon junction complex (EJC) which indicates a splicing event has occurred. During translation, if the ribosome encounters a premature stop (or termination) codon (PTC) and then sees one of these EJC way-points, it signals the corrupted message for degradation.
Several predictions come out of these models including the lack of introns in the 3′ UTR and that the average length of exons should be correlated with the window that the proofreading mechanism can operate on. These are discussed in several papers out of Mike Lynch’s lab including (Lynch and Connery 2003), (Lynch and Kewalramani, 2003), (Lynch and Richardson, 2002) and recently (Scofield et al, 2007).
Efforts to understand the splicing machinery, particularly in S. cerevisiae have led to the discovery of numerous genes that code for proteins that make up the spliceosome. Some of these include small RNAs as well as protein coding genes. The SR proteins are serine-arginine rich proteins that regulate splicing and are found in almost all eukaryotes including most fungi (even those with few introns, such as S. cerevisiae). SR proteins play a role in splicing and in nuclear export (Masuyama et al, 2004, Sanford et al, 2004) indicating that a coupling of these processes may explain why genes with introns tend to be more highly expressed. The evolution of the spliceosomal family of genes is also interesting because the families appear to diversify in some eukaryotes perhaps where there are more elaborate splicing and regulatory action (Barbosa-Morais et al, 2006).
There is some debate as to whether splicing occurs after the pre-mRNA is completely synthesized or if it happens as transcription is occurring. Work on this has shown that both spliceosomal assembly can co-occur with polymerase during transcription, as well as evidence that most splicing (in yeast) is post-transcriptional (Tardiff et al, 2006). It is argued that the two steps occur together to maximize efficiency and fidelity (Das et el, 2006, Moore et al, 2006), but perhaps affinities are species-specific and have evolved to correlate with intron densities?
[Note: This post has links to non-open access journal articles. At this point I am still referring to these even if they are not all readable by everyone, because they contain some data that is only available there. I will strive to focus more narrowly on only papers that are available as open access through pubmed central or directly through open-access journals.]
A recent paper describes the discovery of 9 new introns in Saccharomyces cerevisiae by Ron Davis’s group at Stanford, using high density tiling arrays from Affymetrix. The arrays are designed for both strands allow the detection of transcripts transcribed from both strands. The arrays were also put to work by the Davis and Steinmetz labs to create a high density map of transcription in yeast and for polymorphism mapping from the Kruglyak lab.
Whole genome tiling arrays have also been employed in other fungi. For example, Anita Silâ€™s group at UCSF constructed a random tiling array for Histoplasma capsulatum and used it to identify genes responding to reactive nitrogen species. A similar approach was used in Cryptococcus neoformans to investigate temperature regulated genes using random sequencing clones.
As the technology has become cheaper, it may become sensible to use a tiling array to detect transcripts rather than ESTs when attempting to annotate a genome. In the Histoplasma work transcriptional units could be identified from hybridization alone. Some of the algorithms will need some work to correct incorporate this information, and the sensitivity and density of the array will influence this. These techniques can be part of a resequencing approaches or fast genotyping progeny from QTL experiments when the sequence from both parents is known (or at least enough of the polymorphims for the genetic map).
What is superior about the current Affymetrix yeast tiling array is the inclusion of both strands. This allows detection of transcripts from both strands. Several anti-sense transcripts in yeast have been discovered recently including in the IME4 locus through more classical approaches, but perhaps many more await discovery with high resolution transcriptional data from whole genome tiling arrays.
A paper in Nature this week describes how a few mutations can alter the interactions between species in a biofilm from competitive to cooperative system. This is a great study that goes from start to finish on studying community interactions, looking at an evolved phenotype, and understanding the genetic and physiological basis for the adaptation.
Acinetobacter sp. and Pseudomonas putida were raised in a carbon-limited environment with only benzyl alcohol as the carbon source. Acinetobacter can processes the benzyl alcohol, while P. putida is unable to. Acinetobacter takes up the bezyl alcohol and secretes benzoate that P. putida can then use as a carbon source. The research group propagated these in chemostats and looked at different starting concentrations of the organisms. They found that evolved P. putida had a different morphology and did several experiments to determine the relative fitness of the derived and ancestral genotype.
They went on to also map the mutations in P. putida and found two independent mutations in wapH (I think this is the right gene)â€”a gene involved in lipopolysaccharide (LPS) biosynthesis. They then engineered the ancestral strain to have a mutation in P. putida and found the rough colony phenotype morphology indistinguishable from the strain derived from experimental evolution.
There are various evolutionary and niche adaptation implications arising from this study. One application to mycology is to how lichens evolved in that an algael cell and a fungal cell must communicate and cooperate.
The current contributors to this blog are
Jason Stajich maintains this blog site, a wiki for collaboration, and software and is an Assistant Professor University of California, Riverside in the Department of Plant Pathology and Microbiology and Institute for Integrative Genome Biology. He also provides some genome browsers for fungal genomes as part of his research and in collaboration with the community.
Thomas Sharpton is a postdoctoral fellow at University of California, San Francisco in the Gladstone Institute
Chris Villalta, grad student at UC Berkeley
Balaji Rajashekar was previously at Lund University
We welcome other participants. If you would like to contribute content to this site or to our wiki, please sign up for an account and contact Jason by email.
The JGI has previously released A. niger strain ATCC 1015 sequence in November 2005. ATCC 1015 is used in industrial production of citric acid as it is one of the best producers of citric acid. In Nature Biotechnology a Dutch team has published the sequence of another strain, CBS 513.88 which is an early ancestor of ATCC 1015 used in industrial enzyme production.