I took a turn at lab meeting this afternoon, and used the time to brainstorm about what experiments and analyses we should try to get done before the NIH deadline in Early February. The work for the first NIH Aim will also be preliminary data for a resubmission of the CIHR proposal I just submitted, should that become necessary. I had hoped to also go over other work for the CIHR resubmission but didn't get to it before my time was up, so I'm going to do that next week (remembering to also update the lab meeting schedule).
Goals for Aim 1: Here we have two goals. The big one is to reisolate enough clean DNA from the periplasm that we can either send it for Illumina sequencing (if we have the opportunity to do this - the post-doc has a contact) or characterize it with conventional sequencing, and then do some analysis.
The second goal is to generate and analyze some surrogate sequence data, just to show that we know how to analyze it. By first doing this surrogate analysis, we'll be all set to quickly analyze whatever read sequence data we are able to get before the NIH deadline. Made bold by my recent success in adapting our Perl simulation of uptake sequence evolution to score DNA sequences, I've claimed that I can easily modify the simulation code so it generates sets of DNA sequences from a genome, a random-fragment set and a high-uptake set. Ideally these sets would contain fragments with a range of lengths, but for now I'll just generate several sets of each type (e.g. a 0.5 kb random set, a 1 kb random set set, a 2 kb random set) and mix them together.
Goals for Aim 2: The big issue here is whether we can successfully reisolate input DNA from the cytoplasm. This DNA is expected to be single-stranded, and to be only 5-10% of the total DNA in the cytoplasm (the rest is double-stranded chromosomal DNA). Traditional methods of separating ssDNA from dsDNA rely on their differing physical properties; both hydroxyapatite and BND-cellulose can be used. An alternative I just thought of is to use random hexamers to prime DNA synthesis on the single-stranded fraction of total cytoplasmic DNA, including a biotinylated nucleotide in the dNTP mix. The new DNA can then be separated from the rest using streptavidin agarose or beads, and the template strands eluted by denaturing the DNA.
One complication is that the chromosomes will contain short single-stranded segments, but these are likely to be a small contribution relative to the input DNA. Another complication is that the DNA in the cytoplasm is transient. We'll use a rec-1 mutant so the DNA doesn't recombine with the chromosome, but degradation by cellular nucleases is certainly a concern.
Goals for Aim 3: Here we plan to characterize the results of chromosomal recombination, by deeply sequencing a very large pool of DNA from cells transformed with DNA of another strain. Because there is no DNA reisolation step, I think the best preliminary data we could have is the sequence of one transformed strain (lab strain transformed with DNA from a diverged strain). We can easily get this information from one Illumina lane, and we may be able to get one lane done well before the NIH deadline.
The plan is for the post-doc to first transform a cloned novobiocin-resistance (novR) allele into the most transformable of the three sequenced diverged strains. This can be done very quickly because the former post-doc left a tube the novR DNA, and the transformation can be done by overnight co-culture of cells and DNA. The next step is to grow up and prepare DNA from one of the novR transformants, and to use that DNA to transform the standard strain KW20. Then we (i.e. the post-doc) just need to prep DNA from one of the transformants and pass it on to the Illumina specialists. They will shear the DNA to the appropriate length for Illumina sequencing, ligate the Illumina adapters onto the ends, and sequence it for us. We can use this sequence to specify the expected density of recombinant segments in the pooled chromosomes we propose to sequence, and the length distribution of recombination tracts.
Leroy Hood and the tool-driven revolution in biology
1 day ago in The Curious Wavefunction