Hi,
When I process a cram file on my server everything works fine, no errors. When I do the same on Travis CI I get the following:
[E::cram_get_ref] Failed to populate reference for id 0
[E::cram_decode_slice] Unable to fetch reference #0 1..2100774
[E::cram_next_slice] Failure to decode slice
I reasoned that this is the case because it cannot find the reference fasta with which the cram file was made, so I included the reference in the test dataset.
However, the path in the cram header is an absolute path on my computer, and the path on Travis CI is obviously not the same.
I guess if I could specify in pysam which reference fasta to use everything would be fine? I guess pysam/samtools attempts to fetch the reference fasta.
Or should I manually adapt the cram header to get a relative path to the reference?
Do you have suggestions on how to fix this?
Cheers,
Wouter