Sean

With `--symlink`, the performance seems fine. So probably not urgent so support something like this. Thanks.

I think this happened with my pull request https://github.com/biopython/biopython/pull/3733 also. There were two CI errors, but neither of them seemed to be at all related to the code I wrote....

> Why are you removing all those warnings anyway? That's a good question. Actually I didn't mean those last few commits to be part of the pull request (didn't realize,...

avoiding the regexes is helpful for speed. The changes in that commit are all the result of profiling I did. All together the changes almost halve the parsing time (9...

I made a commit reverting changes irrelevant for this pull request (except the one changing the quotes regex to checking the first and last characters). Sorry for the confusion.

Ok, I tried. The history looks even more complicated now, so not sure if I did it right.

I'd love to see this option also, particularly in Foldseek (which I guess is somehow downstream of MMseqs2?).

> We've been thinking about contributing some higher level "neighbourhood" functions to Biopython along a similar line. Basically a way to say "give me 20 kbp upstream and downstream of...

@BjornFJohansson I found the code for `shifted` in the pydna repository. It is very helpful. Thanks! It would also be great to have something like this integrated into Biopython.

Interestingly, when I run: `foldseek easy-search "query.pdb" UniProt results.html tmpDir --threads 16 --prefilter-mode 1 --split-memory-limit 20G --format-mode 3 --max-seqs 100 -e 0.001` The execution time halves from 1 hour to...