Jonathan Pevsner received his Ph.D. in Pharmacology and Molecular Sciences at Johns Hopkins (1989), and pursued postdoctoral training at Stanford University in the Dept. of Molecular and Cellular Physiology. He joined the faculty of the Department of Neurology at the Kennedy Krieger Institute in Baltimore where he is currently an Associate Professor and supervises a research lab that studies the genetic basis of childhood brain disorders. Dr. Pevsner began teaching courses in bioinformatics and genomics at Johns Hopkins in 2000, and he won Teacher of the Year awards twice in the School of Medicine (2001, 2006), once at the Homewood campus MLA program (2009), and he was given the Professor's Award for Excellence in Teaching (2003).
The Bioinformatics and Functional Genomics book that he authored is among the best introductions to the field. The book is accompanied by a well maintained website that features a complete bioinformatics teaching curriculum http://www.bioinfbook.org/
Dr Pevsner is currently preparing the third edition of his book that will also include topics on Next Generation Sequencing technologies.
What hardware do you use?
We use a nice homegrown cluster that includes head nodes, three compute nodes (each having 128 GB RAM and ~100 cores in all), and several hundred TB of storage. Much of our data is from patients and their families, so I prefer to use a local cluster to avoid security concerns about placing any protected health information on external servers (e.g. cloud computing).
What is your text editor?
Vim. For teaching I suggest nano. On a PC I like Crimson Editor, and TextEdit on a Mac.
What software do you use for your work?
We use software in different areas. For analyzing next-generation sequence DNA data it's often aligners such as BWA-MEM; variant callers such as GATK and somatic variant callers such as Strelka and MuTect; for variant prioritization we use VAAST. (When Matt Shirley was a grad student in my lab he used Strelka and VAAST to find a variant that causes a rare disease)( http://www.ncbi.nlm.nih.gov/pubmed/23656586 ).
For RNA-seq we use Bowtie, Cufflinks, RSEM, CuffDiff, and Ballgown.
For analyzing genetic relatedness we use our own software (kcoeff, snpduo, SNPtrio, triPOD) and packages such as PLINK and BEAGLE. We're collaborating with Partek to develop additional tools on their platform. For phylogeny I often use the suite of tools in MEGA software. At NCBI in addition to BLAST my favorite tools are PheGenI (Phenotype-Genotype Integrator) and EDirect (command-line access to Entrez databases).
What do you use to create plots and charts?
We use R, and we use Partek very heavily to visualize all sorts of data sets.
What do you consider the best language to do bioinformatics with?
Python is better than Perl. But the best language depends on the particular task you're trying to solve.
What bioinformatics tools/software do not get enough recognition?
Don Freed, a graduate student in my lab, thinks that BWA-MEM for alignment and LUMPY for structural variant detection are fantastic packages that have not yet been recognized enough.
See all post in this series https://www.biostars.org/t/uses-this/
To be notified of new post in the series follow the first post: Jim Robinson of the Integrative Genomics Viewer (IGV) uses this