Chris Miller

We currently have several openings at TGI. One fis or a statistically-oriented staff scientist, and two for more traditional software developers (though bioinformatics experience is a plus!). Though the job descriptions are often boring, the work is anything but! We're doing kick-ass genomic research on cancer and other human disease and building software pipelines designed to make biological insights from petabyte-scale data. Apply through the links below, but feel free to message me with questions and I'll either answer them or put you in touch with someone who can. Staff Scientist https://jobs.wustl.edu/psp... Software Developers (2) https://jobs.wustl.edu/psp...... - Chris Miller

Mark Gerstein is putting together a list of US programs in bioinformatics here and is soliciting additions: http://blog.gerstein.info/2013... I figured that if anyone can help him expand that list, it's the Biostar community. - Chris Miller

Someone pointed me to Drake the other day. I haven't used it, but it seems appropriate for a lot of bioinformatics tasks: http://blog.factual.com/introdu... - Chris Miller

Just be careful - don't want to discourage either new bioinformatics folks or ESL users - Chris Miller

FYI, a spam post still got through to the RSS last night. (id 62389). Just another data point, in case you hadn't noticed - Chris Miller

I feel like the 4-hour delay is a little long. In the interest of timely answers, can we cut it to 2 hours? Do we have enough international coverage with our moderators to cover spam deletion that quickly when it crops up at odd hours? - Chris Miller

Moderator approval of the first post seems reasonable to me, as long as it can trigger an email notification or something. I'm happy to approve some posts if it means cutting down on this nonsense. If Istvan's new countermeasures don't work out, we could give it a shot. - Chris Miller

Right. This calculation used single-end reads, so the numbers will be lower than what you can get from paired-end reads, using that extra information. - Chris Miller

Breakdancer doesn't call copy number variants. It calls structural variants. If you see a large deletion in breakdancer, it may be the case that a copy number deletion has occurred. It's also possible that the deleted sequence got re-integrated somewhere else, in which case, there would be no effect on copy number. - Chris Miller

It may depend on the source, but in my experience, it's defined as the percentage of bases that are incorrectly called. The 0.8% error rate that you describe would mean that of every 1000 bases coming off the sequencer, 8 of them will report the incorrect base. - Chris Miller

I'm not sure I entirely understand your question, but there doesn't appear to be anything wrong with the output you posted. It indicates that you have 7 mutations in LATS1, all SNVs. If you had 2 indels and 2 SNVs, the total mutations would still be 7. According to the FDR values, this gene is mutated more frequently than would be expected, based on the background mutation rate in the samples. - Chris Miller

There isn't any single answer to this question. It all depends on what kind of biological question you're trying to answer with the RNAseq data. If I'm looking for differential exon usage due to spliceosome mutations, gene-level data is useless to me. If I'm trying to work with a huge network of genes, I may need to simplify my inputs and use gene-level metrics to make the problem tractable. - Chris Miller

No worries, man. If I had a dollar for every time something turned out to be way harder than I expected... - Chris Miller

a) I don't know of any off-the shelf tools, b) this is generally difficult, because breakpoint-spanning reads may map to one side or the other, but also may not map at all. To get a high confidence call, you could create a short contig containing your breakpoint sequence +/- 200 bp, append it to the reference genome, then realign all of your reads against this. Compare the depth of breakpoint spanning reads on your contig to the depth at the breaks on the original reference sequence, and that ratio will give you a pretty good idea of the frequency. - Chris Miller

Even if it was a missing zero typo, 3200 a month would be 38,400 a year, which is less than a postdoc salary. - Chris Miller

I'm assuming POA = "Plan of Attack". You're going to have to be a lot more specific about what you're asking in order to get any help here. I'm closing this post. If you have a specific question about some aspect of gene expression analysis, feel free to try again with a new question. - Chris Miller