gbouras13/single_marker_gene_microbiome_analysis

Repository to hold analysis for single marker genes in HNSC Microbiome

single_marker_gene_microbiome_analysis

Repository to hold analysis for single marker genes in HNSC Microbiome

• Look at 16S and rpob dnak genes, especially the single copy genes - rpoB has the most literature
• Add any paper links to this repoistory

http://frogs.toulouse.inra.fr

Bioinformatics Ideas

1. Create a list genera/species known to be in the HNSC microbiome (in progress). Err on the side of completeness

• eHOMD database for human oral microbiome database
• GTDB identification/Silva identification/ Actual name we want (species name) - big spreadsheet

2. Pull all sequences from GTDB https://gtdb.ecogenomic.org especially dnak and rpoB gene - MLST typing probably helpful here

• To be done - how to actually get 16S from Silva https://www.arb-silva.de and rpoB genes from GTDB
• shell scripts to do this (gtdb has own program I think)
• Hard part - getting rpob genes. Will likely involve prokka -> extraction of gene. Might create my own program eventually see below
• Should be handled with snakemake (George) when actually run, but for now, Kenny figure out how to do it on 1 sample (python).

3. Create multiple sequence alignment (MAFFT/PRANK) of the genes and idenity regions that are conserved

• Very easy once this 2 is done.

4. Create primers - LOGO to do this

Other ideas (George)

• Try to apply phylomagnet to TCGA
• rpob gene extraction program? like barrnap

Other notes from Meeting with Przemeck

• 5-8 nts at the 3' end are the most important and must be there to amplify
• Likely to use EUPA code (ambiguity in some spots of primer)
• Consider orthoANIu instead of FastANI as it gives coverage
• SRA search for other WGS databasets
• Search for antibiotics in the MAGs

Other literatue

• https://bmcmicrobiol.biomedcentral.com/articles/10.1186/s12866-019-1546-z
• https://www.mdpi.com/2073-4425/11/2/139/htm

Plan

• Everything saved as shell scripts in this directory for now in this directory (.sh files)
• Do not upload genomes (too big)