Create-Marker-Gene-Database

Scripts for creating "marker gene" reference files for use as database for sequence similarity searches

Overview

The generate_annotation.sh is taking a FASTA file with common header format as can be usually found in UniProt, NCBI, etc. and performs the following:

• Dereplicate sequences by 97 percent similarity with cd-hit
• Convert FASTA to one-line format and remove header description (after whitespace) with awk
• Convert headers to md5 with python script convert_fasta_headers_md5.py
• Merge files to create a complete annotation file with python script merge_annotation_files.py

Software requirements

cd-hit
Pandas (Python)

Requred files

• FASTA file (.fa extention)
• Tab-seperate annotation file (same name as FASTA with .tab exention)

Downloaded from UniProt, for example:
https://www.uniprot.org/uniprot/?query=terminase+family%3Aprotein+taxonomy%3A%22Viruses+%5B10239%5D%22&sort=score

Output files

1. derep.fa
2. derep.md5.fa
3. derep.fa.clstr
4. annotation.tsv