OMA Standalone: "could not read the right number of characters"

Question

[inactive account]

I'm trying to run OMA Standalone (version 2.2.0, darwin x64) on a few nucleotide FASTAs. It creates caches for several files without error (i.e, the .db, .map and .tree files), but fails on others with: Error, (in ConvertGenomes) Could not read the right number of characters.

It's such a cryptic message, and I could not find the source code for the tool, so it's impossible to divine what it stumbles on. Could the devs or people with experience with OMA standalone shed some light?

Thanks

[GenoMax]

tagging: adrian.altenhoff

[Adrian Altenhoff]

Hi, this is likely related to some softlinks in input data or cached databases. could you retry to rerun with -d 10 and let us know about the error. thanks! Adrian

[inactive account]

Hi! I replaced my symlinks with copies of target files, ran OMA with -d 10 and got this:

gname := homo_sapiens
genomes := [danio_rerio, gallus_gallus, homo_sapiens]
<-- error in ConvertGenomes = Could not read the right number of characters}
Error, (in ConvertGenomes) Could not read the right number of characters
        executing statement: checksum := sha2(ReadRawFile(g))[1..16]
        locals defined as: dataset_hash = d67b9f6420c7b496154f02753144278935476\
2f374e369933284b7c0ab3f6bdd200ef3b1500b9e92fe8d7d85a7daba18e4df402984a8d88a76e1\
b0b0f33c7f6c, g = DB/homo_sapiens.fa, off = 12, gname = homo_sapiens, checksum =
a2eae7aea891cbfd, spliceFn = DB/gallus_gallus.splice, fcheck = a2eae7aea891cbfd,
code = code, input = input, ML = ML, splicemap = splicemap, setNr = setNr, step
 = step, start = start, end_pos = end_pos, i = i
        ConvertGenomes called with arguments: [DB/danio_rerio.fa,
DB/gallus_gallus.fa, DB/homo_sapiens.fa, DB/macaca_mulatta.fa,
DB/papio_anubis.fa, DB/rattus_norvegicus.fa], ParallelInfo(1,1)

[Adrian Altenhoff]

It looks like the fasta file of DB/homo_sapiens.fa is corrupted. Does it contain any non-unicode characters or ascii char 0 in it? you can also send me this fasta file if you are unsure how to check. the email adresse would be contact at omabrowser.org.

[djw]

Hi there,

Was there any resolution to this issue? I'm curious as I'm getting the same error as the OP, but following Adrian's advice here, I cannot find any illegal characters in any of my fasta input files. Many thanks, djw

[Adrian Altenhoff]

Hi djw, no idea, kirill.grigorev did not get back to me - so I assume it was this problem. if for you this is not the case would you mind sharing the dataset (or parts where the error happens) with me? you can reach my by email on contact at omabrowser org.