- annotate all the assemblies with prokka and cluster the genes in groups of orthologs using roary
with -i flag in roary you can decide the minimum percentage identity for blastp (default 95%); do you think this value is appropriate for your dataset?
Note: since your dateset is very small for performing GWAS, you will select a set of few accessory genes to be used in the final part of the course
- from gene_presence_absence.csv produced by roary select three accessory genes, present in a significant number of samples (e.g. 1/3 of your dataset)
- add to the tab-separated values (TSV) file produced during working with your data part 2 session 2 three extra columns corresponding to the three selected genes, as follow:
FILE ResistanceGenes Gene1 Gene2 Gene3
Samplename_1 GeneName-GeneName-...-GeneName present absent present
Samplename_2 ND present absent absent