Wow very detail, Thanks! I will give it a try.
Hello everyone, I am in the process of annotating my genome and am at the final step of filtering the annotated gene structures. I have noticed numerous papers referencing the final step of filtering genes based on varying coverage overlap with transposable elements (TE), such as 50%, 90%, and so on. I would like to perform a similar step but am uncertain about the procedure. Could anyone offer some guidance? Thank you!
1 answer
One option is to use BEDtools to find out how much your genes overlap with repetitive DNA. It would help if you have a GFF of repeats in the genome, similar to what RepeatMasker provides.
awk '$3=="gene"' genes.gff | bedtools coverage -a - -b repeats.gff > genes_repeat_cov.gff
Then select genes that overlap based on your threshold (e.g. 50%):
awk '$NF>0.5' genes_repeat_cov.gff
Then you can provide the IDs of the selected genes to agat_sp_filter_feature_from_kill_list.pl to remove them from the original GFF:
agat_sp_filter_feature_from_kill_list.pl --gff genes.gff --kill_list genes_to_remove.txt --type gene
Log in to answer this question.