Visualize variant data.

Written by Tom Sasani for MDCRC 6521 at the University of Utah.

nanoplot is a very basic toolkit that allows users to create useful visualizations of VCF metadata in bokeh.

During the course of a rotation in Aaron Quinlan's lab, I performed a number of whole genome sequencing experiments on a strain of Vaccinia virus. Once I obtained sequencing reads, I aligned them to the Vaccinia reference genome and found mutations in my sample with a program called Freebayes. The full list of variants present in the virus samples I sequenced is stored in a "variant call format," or VCF, file. Because the Vaccinia genome is relatively small (~190 kilobases), the VCF files I produced only contained a few hundred to a thousand individual variants, including both single and multi-nucleotide polymorphisms. As a way to visualize the data in these VCF files, I wrote nanoplot, a simple program that plots each variant as a function of its quality score (the likelihood that the variant is "real," and not called in error).

To visually determine which variants in a VCF meet a quality score threshold, simply run:

python nanoplot.py -v [VCF] -t [threshold] -vtype [variant type]

The -vtype option should be followed by either 'COM' or 'SNP', depending on whether you want to visualize complex variants (i.e., multi-nucleotide polymorphisms) or SNPs.

nanoplot requires Python 2.7. Additionally, it requires the following two libraries.

bokeh ____To install, run conda install bokeh.

cyvcf2 ____To install, run pip install cyvcf2.