Comparative genomics axtChain, potential issue.

[plattsad]

Hi. I just wanted to flag a warning about the comparative genomics pipeline and specifically this line:

axtChain $axt $ref $query $chain -linearGap=loose -faQ -faT

I was using a very similar Kent-like pipeline for some comparative work and discovered that the -faQ and -faT options
gave different results than those obtained using 2bit files. It looked like the chained alignments generated using
the -faQ and -faT options were often of poorer quality.

After discussing this with the UCSC genome browser folks it turned out that there was indeed a bug in axtChain and
this was fixed in May of 2023 (so probably only in June 2023 onwards releases):

ucscGenomeBrowser/kent@7106942?diff=split

Prior to this, results obtained from axtChain in -faQ -faT mode may have been problematic.