I am a bioengineering undergrad at UC Berkeley, and I spent the Spring and Summer of 2011 doing projects for the Holmes Lab.
Created a Galaxy workflow for visualizing viral genome secondary structure annotations in JBrowse.
The input for this pipeline was a set of unaligned, homologous viral sequences, and the output was a JBrowse window displaying (1) the most likely secondary structure of the sequence, and (2) the relative likelihood of each individual base forming a base pair. Essentially, the procedure was: align sequences using FSA
-> construct a neighbor-joining tree with xrate -> use the alignment and tree, along with the pfold grammar, to predict secondary structure information with xrate -> display secondary structure annotations in JBrowse.
Wrote documentation for the JBrowse server-side API on gmod.org.
In addition to the documentation for JBrowse-1.2.1, accessible from the jbrowsedev/current page, I also provided a section for documentation of changes and new features that have yet to be released: jbrowsedev/upcoming.
Updated the hg19 demonstration on jbrowse.org.
For applicable tracks, features are now searchable by name and are linked to annotation pages.
Produced phylogenetic trees for glucuronyltransferase sequences.
In collaboration with the Anderson Lab, I produced alignments and trees for sequence sets containing human and bacterial glucuronyltransferase homologs. FSA
was used as the alignment program, columns containing >90% gaps were removed, and xrate was used to make neighbor-joining trees.
Created a simple Galaxy workflow for an ancestral reconstruction analysis.
The input was a set of unaligned fasta protein sequences, and the output is a sequence alignment of all input sequences and their predicted ancestral sequences. The procedure was: align sequences using FSA
-> eliminate columns containing >90% gaps -> produce a neighbor-joining tree from the edited alignment using xrate -> predict ancestral sequences using the alignment and tree as inputs to protpal.