Jeff Leek's non-comprehensive list of awesome things other people did in 2013

Jeff Leek, biostats professor at Johns Hopkins and instructor of the Coursera Data Analysis course, recently posted on Simly Statistics this list of awesome things other people accomplished in 2013 in genomics, statistics, and data science.

At risk of sounding too meta, I'll say that this list itself is one of the awesome things that was put together in 2013. You should go browse the entire post for yourself, but I'll highlight a few that I saved to my reading list:

• Yuwen Liu and colleagues published a paper on power for RNA-seq experiments comparing biological replicates and sequencing depth. This paper adds to previous work showing conclusively that adding more replicates is usually better than adding more sequencing depth.
• Thomas Lumley's brief but insightful discussion of statins and the causal Markov property. Lumley also linked to a Labhacks post on a high-quality scrunchable cloth poster print for $25.
• Joe Blitzstein's Data Science course at Harvard, which has an excellent reading list (many are IPython Notebooks, which can be read freely and downloaded to run locally.
• A link to Lior Pachter's blog, which I also regularly read for all things related to RNA-seq and beyond.
• Tuuli Lappalainen's paper on RNA-seq reproducibility. You can also follow Tuuli on Twitter (@tuuliel).
• On the metagenomics front, the metagenomeSeq R package for differential abundance in microbial communities from Mihai Pop's group, and from Curtis Huttenhower's group, a new method for predictive functional profiling of microbial communities using 16S rRNA marker gene sequences.
• Finally, because Jeff's post was about awesome things other people did in 2013, he deliberately omitted some of the awesome things that his own group and others at Johns Hopkins did in 2013. Some of Jeff's contributions that I found most useful in 2013 were several guides, available as GitHub repository READMEs:
• jtleek/rpackages: A guide to developing R packages using devtools, roxygen2, knitr, git, and GitHub.
• jtleek/datasharing: A guide to how to share data with a statistician. This is a great post to send to your collaborators who constantly send you inconsistently coded data without a metadata code book.
• jtleek/reviews: A guide to reviewing academic papers. We've all gotten back terrible, often times useless reviews. Follow the advice here to avoid being one of those awful reviewers we all despise.

This only a sample of what's posted on Jeff's blog. Go read the full post below.

Simply Statistics: A non-comprehensive list of awesome things other people did this year.

Curoverse raises $1.5M to develop & support an open-source bioinformatics data analysis platform

Boston-based startup Curoverse has announced $1.5 million in funding to develop and support the open-source Arvados platform for cloud-based bioinformatics & genomics data analysis.

The Arvados platform was developed in George Church's lab by scientists and engineers led by Alexander Wait Zaranek, now scientific director at Curoverse. According to the Arvados wiki:

Arvados is a platform for storing, organizing, processing, and sharing genomic and other biomedical big data. The platform is designed to make it easier for bioinformaticians to develop analyses, developers to create genomic web applications and IT administers to manage large-scale compute and storage genomic resources. The platform is designed to run on top of "cloud operating systems" such as Amazon Web Services and OpenStack. Currently, there are implementations that work on AWS and Xen+Debian/Ubuntu. ... A set of relatively low-level compute and data management functions are consistent across a wide range of analysis pipelines and applications that are being built for genomic data. Unfortunately, every organization working with these data have been forced to build their own custom systems for these low level functions. At the same time, there are proprietary platforms emerging that seek to solve these same problems. Arvados was created to provide a common solution across a wide range of applications that would be free and open source.

A few questions should be apparent: What value does Arvados provide over other more widely used platforms (namely, Galaxy) that also aim to enable reproducibility, transparency, sharing, collaboration, and data/workflow management with biological big data? And how does Curoverse distinguish itself from other cloud-based bioinformatics services like Seven Bridges, DNA Nexus, and the next implement-GATK-on-Amazon-and-sell-it-back-to-me service provider that pops up? I understand that there are real costs with free software, but will the service that Curoverse provides be valuable and cost-effective enough to overcome the activation energy and make up for the "switching costs" that the average bioinformatician faces on adopting a new way of doing things? While the platform and the support model sound potentially very useful, these are all questions that the Curoverse team will need to carefully consider in attracting new users.

Arvados open-source bioinformatics analysis platform: https://arvados.org/

Curoverse: https://curoverse.com/

Press Release: http://www.prweb.com/releases/2013/12/prweb11424292.htm

Biostar Tutorial: Cheat sheet for one-based vs zero-based coordinate systems

Obi Griffith over at Biostar put together this excellent cheat sheet for dealing with one-based and zero-based genomic coordinate systems. The cheat sheet visually explains the difference between zero and one-based coordinate systems, as well as how to indicate a position, SNP, range, or indel using both coordinate systems.

Biostar Tutorial: Cheat sheet for one-based vs zero-based coordinate systems