Comparative genomics and phylogenetics pipelines for the Zamanian Lab.

Every comparative genomics project should be encapsulated in a separate directory in this repository (i.e., GitHub/Phylogenetics/AmineR, GitHub/Phylogenetics/BetaTubulin). Each investigator is welcome to organize that directory as they see fit, but generally it should contain the following steps:

1. Gene identification
   • This includes the entire pipeline for creating a filtered set of related genes.
   • See AmineR/get_homologues.R and PPN_ChemoR/identification_pipeline.sh for two examples, the first of which uses the WormBase ParaSite orthology calls and the REST API, and the second that uses seed sequences from C. elegans and a reciprocal HMM/BLAST approach.
2. Alignment and tree-building
   • There are a number of potential alignment strategies, and these pipelines should be organized in a single script. The ChemoR pipeline first built a scaffold of HMMs using curated C. elegans sequences, and then it used MAFFT to sequentially add parasite genes to this scaffold. The AmineR pipeline trimmed sequences to only include transmembrane regions and aligned these.
   • The multiple sequence alignment probably needs to be trimmed (removing uninformative columns and gap-heavy sequences). This can be performed manually using an editor like Seaview, but it can also be done automatically with trimal. Note that the parameters used for trimming will need to be heuristically selected.
   • After trimming the alignment to informative regions, the tree is inferred. Two general approaches are preferred, either maximum-likelihood (ML) or Bayesian approaches. For ML, use either RAxML or IQ-TREE. For Bayesian, use MrBayes.
3. Tree visualization and annotation
   • After finishing the tree, the lab uses the R package ggtree to visualize and annotate the consensus tree. See AmineR/aminer_tree.R for an example script.

It is recommended to include a Markdown file (notebook.md) in each project directory that is a running archive of approaches that were developed, attempted, discarded, and eventually selected.

All output files should be written to /Box/ZamanianLab/Data/Genomics/{Project}. Files should be separated by species until the required concatentation prior to alignment. It is recommended that the output of each step of the identification pipeline is written to a directory that can be organized in ascending numeric order (see /Box/ZamanianLab/Data/Genomics/AmineR) as an example. The output of the tree inference (most likely run on the BRC6 server) should be copied into this directory. It is then recommended to create a separate ggtree directory, where output trees can be written in ascending numeric order (see /Box/ZamanianLab/Data/Genomics/AmineR/phylo/4/ggtree).