Let’s say im using naive LCA.
a read is :-
Species A 100%
Species B 100%
Species B 100%
Species B 100%
Species B 100%
at the moment if ptc is 100%, it goes to lca(spA,spB), but if i reduce ptc to 51% it goes to spB. this is what im finding currently. is it simply a numbers game in this example? ptc 51% takes the top 51% of matches and looks at which hit the most?
@OmarKR, this shouldn’t be the case. Are the proteins all the same length? If the PTC is 100%, it might cover proteins that are short and fully covered. At 51%, it will also consider the longer ones.
Another thought: if the small-length protein from Species A and a protein from Species B both have 100% PTC, only these two proteins are involved. However, at 51% PTC, other proteins from Species B come into play, and the protein from Species A should also be covered in this scenario.
Alternatively, since we only look at 25 alignments per read, Species A’s protein might not be included with 51% PTC because more proteins from Species B could come into play, potentially having higher bit scores than the protein from Species A.
Can you send me a screenshot of the inspector window? A screenshot of the alignments in the inspector for both cases would help us understand what’s going on. Right-click on the node, select “Inspect” from the drop-down list, select the read of interest, expand its alignments, and send us the screenshot (it should show 25 hits for a read).
Regards,
Anupam