Hi, I carried out BLASTp using a local version of BLAST for ebola and zaireebola to look for orthologous genes. Are orthologs all those genes that have a percent identity above 100%? How can I find the sequence similarity between the orthologs?
If your output is in regular text, the value is defined in the alignment section, if you used tabular or other output, you need to specify the ppos value to be reported, which is the percentage of positive values.
if you are specifically looking for orthologous genes then blast is only the first step. You should do some post-processing of your blast result to get to define orthologous relationships.
I'm thinking of tools such as, inparanoid, orthofinder, ... ,they take the blast result as input. If the orthology relations is not super critical you can get away with 'lower level' approaches such as best-bidirectional hits (blast A vs B, then B vs A and check for hits that are best ones in both directions) or apply something like the Li-Rost criterion (== get hits that have aligned region over 100-150 long, with >30% identity (not similarity/positives) ! )
EDIT: have a look here for more info: What Is The Best Method To Find Orthologous Genes Of A Species?
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
I do have orthologs in the output. I need to find a way on how to find the sequence similarity between those orthologs. Is there any way for me to find the similarity. I used a local version of BLAST. My output includes percent identity, evalue, bit-score and query coverage.
getting confused here ... what do you mean with:
if orthologous genes are present in the dataset then indeed, they will be present in your blast output (trick is to get the right relations though). Alternatively, do you mean that you know the orthologous relationships already and want to get some stats about them from the blast output. If the latter than have a look at the answer of JC