Not surprisingly, the post-doc has a more ambitious collection of projects than the new grad student does. He's on the other side of the continent visiting our collaborator and learning how to set up a whatchamacallit (an assembly line for analyzing data...). I'll see if I can summarize all his projects.
His first project is the analysis of recombination tracts. This relies on full-genome sequencing of individual recombinants between our standard lab strain Rd (the recipient) and a clinical H. influenzae isolate that differs at about 2% of alignable positions and by several thousand indels and other structural variations. The first paper, describing analysis of the first four recombinants, is ready to be submitted (tomorrow?), and the multiplexed sequencing of another 60 or so strains should be producing its first results this week (done by the Genome Sciences Centre under our Genome BC grant). (There was a concern about the quality of the libraries, so one lane (8 genomes) is being done ahead of the others to check this.)
His second project is the analysis of recombination frequencies across the genome, for every position that differs between the same two strains. We originally planned to include this with the Genome BC work, but UBC's Biodiversity Centre has a new sequencer and is offering longer cheaper sequencing runs that would be very well suited for this project. BUT, to use their service we'd have to make our own libraries (i.e. the post-doc would have to make his own libraries, or persuade the collaborator's technician to make them for him). It might be better to just do the first pass with the Genome BC/Genome Sciences Centre as planned. But I think the experiments to prepare the DNA that will be sequenced haven't even been started yet. In principle these should be straightforward extensions of the work he's already optimized, but I'm not counting on that.
His third project is the 'QTL-mapping' project, using identification of donor sequences in recombinants between the two strains to map the loci responsible for phenotypic differences. The first phenotypic difference he's looking at is the >1000-fold difference in transformability between the two strains; the second will be (we hope) sensitivity to killing by serum. He's doing this with help from an undergrad (almost-grad), who has been developing the screening method that will identify recombinant strains with reduced transformation. Then now have an assay that should work and that is only moderately more troublesome than what we originally planned. They're going to screen the same strains that are being sequenced for the first project, so here's hoping they find some transformation changes.
His fourth project is the analysis of uptake specificity. The first step of this was very successful, and he's subsequently put a LOT of work into developing methods to analyze the data. We won't do any more sequencing on this project until we get our new DNA uptake grant (or until the reviewers reject it but advise doing additional work). I'm eager to get the present results written up and published asap, but he keeps finding new analyses that he thinks need to be done.
Boundary value conditions, domain applicability and "American Sniper"
7 hours ago in The Curious Wavefunction