This repository includes all of the code related to a project I worked on as part of a first-year lab rotation in the Delwiche laboratory.
The project was inspired by a 2009 study by Lang et. al. that, among other things, looked at correlations between increases in the Transcription-Associated Protein (TAP) family diversity and several milestones in plant evolution such as the development of land plants and the radiation of angiosperms. While the Lang et al. study included genome data from a large number of plant species, there were several lineages for which no data was available, including the Charophyte green algae.
Species studied:
- Cephaleuors parasiticus
- Mesostigma viride
- Nitella mirabilis
- Phaeophila dendroides
- Pyramimonas parkeae
- Spirogyra pratensis
The goal for this project was to extend the analyses done by Lang et al. to include additional green algae datasets, including several Charophyta species.
An additional aim of the project was to determine the feasibility of using expression (RNA-Seq) data to measure TAP diversity.
The original study included only species for which complete genome sequences were available. For this project, we had not genome sequences, but large transcriptome data-sets. The obviously has some major implications for comparing the results of this project with results from the previous study, and also comparing different species within the project.
The result is that we can only make claims about the minimum number of TAP families present in a given species.
RNA-Seq data used for this project was assembled de novo using Trinity.
The first step in the project was to find RNA-Seq contigs containing protein domains found in TAPs.
Profile Hidden Markov Models (HMMs) were downloaded for each of the TAP domains listed in the Lang et al. study. Many of the profile HMMs were downloaded from the Pfam database, along with several custom profile HMMs created by Lang et al.
In the several years that have elapsed since the original study, a number of Pfam domains identifiers have changed, and it was necessary to map the ids used in the original study to the new identifiers:
DUF1004 -> GAGA_bind
DUF246 -> O-FucT
DUF623 -> Ovate
DUF833 -> NRDE
IWS1_C -> Med26
MED6 -> Med6
MED7 -> Med7
SOH1 -> Med31
Old names are listed on the left side while the newer Pfam identifiers are on the right side.
HMMER](http://hmmer.janelia.org/) was used to scan for protein domains in the RNA-Seq contig data.
- Decide on how to deal with multiple-classification contigs