And that’s why we need a human…

So my plane heads out in about 8 hours so I thought I would leave you all with something to consider as you prepare for the workshop in the coming two weeks…

Martin Krzywinski did a really great presentation entitled “Needles in stacks of needles” which he drew from a paper by Cooper et al., 2011 where the authors discuss (Taken from Abstract): The difficulty  in pinpointing the few phenotypically causal variants among the many variants present in human genomes, particularly for rare and complex traits wherein genetic information alone is often insufficient. The authors go on to present a review approaches to estimate the deleteriousness of single nucleotide variants (SNVs), which can be used to prioritize disease-causal variants and describe recent advances in comparative and functional genomics that enable systematic annotation of both coding and non-coding variants.

Martin continues on in his presentation to talk about the challenges of understanding the genetic basis of disease and all the new ‘toys’ that are out there both in terms of GUI programs and coding that allow us to dig into the data, he breaks down via schematics on MAQ, Bowtie, considerations on genome compilation and comparison, mutation profiles and despite the beauty of output from programs like Hive Plots etc…how it’s possible to get into ‘visual jargon’. He goes into the challenges of interpretation and what can go wrong when things are incorrectly encoded.

credit: Drew Shenman, New Jersey - The Newart Star Ledger
credit: Drew Shenman, New Jersey – The Newark Star Ledger

He goes on to highlight more tools that are being used in the field to look at causality and correlation and he touches on many subjects you will hear about in the workshop…

But beneath all the computers, the data crunching, command-line fu-ing and ‘cool’ programs that you will be gaining experience with, don’t forget the human component, don’t lose perspective in this bioinformatic era of ‘overwhelming data’ and mass computation in the field of biology.

454, Ion Torrent, Ion Proton, MiSeq, HiSeq…

Transcriptomics, metagenomics, proteomics…

Whole genome sequencing, single-cell sequencing, exome sequencing…

RNA-seq, ChIP-seq, SMART-seq…

There is a lot of data out there:

Picture2But data without context, insight and intuition of human biological knowledge; computers and fancy programs can only take us so far…

Picture3We’ve got some great humans at Evomics 2014 to get you started…

I highly highly highly recommend you take a spin through Martin’s presentation, it’s a great slide set that doesn’t require a lot of ‘context’ to ‘get’ the slides; he does a great job with it and I think it’ll be a great primer for what you will encounter and the perspective to keep in mind as you travel and prepare for a very packed two weeks.

needles-krzywinski-icdm2012-hires (if this link doesn’t work, try the web link: http://mkweb.bcgsc.ca/icdm2012/needles-krzywinski-icdm2012-hires.pdf)

I also recommend Cooper’s paper.

The fun begins in a week…see you all there

…Dr. Mel

Leave a Reply

Your email address will not be published.