IndexError exception raised due to reads.fas being empty

[terrycojones]

I run shorah.py as follows:

$ shorah.py -b alignment.bam.sorted -f data.fasta

and get an error:

Traceback (most recent call last):
  File "/opt/shorah/bin/shorah.py", line 139, in <module>
    alpha=args.a, keep_files=args.k, seed=args.S)
  File "/opt/shorah/lib/python3.5/site-packages/shorah_dec.py", line 403, in main
    r = list(aligned_reads.keys())[0]
IndexError: list index out of range

The code is assuming that reads.fas contains reads, but in my case that file is empty.

It's not immediately clear to me where reads.fas is created. In shorah.log I see

$ cat shorah.log
INFO 2017-12-03 18:42:07,585 <module>                          130 /opt/shorah/bin/shorah.py -b alignment.bam.sorted -f data.fasta
DEBUG 2017-12-03 18:42:07,585 <module>                          136 running dec.py

Ah, now I see it:

$ strings /opt/shorah/bin/b2w | grep reads\\.fas
reads.fas
w-%s-%d-%d.reads.fas

Any suggestions?

Thanks!

[ozagordi]

Hi! I would first rename your bam file to something like alignment_sorted.bam. Then, are reads correctly aligned with a good coverage? You can inspect the alignment with samtools tview alignment_sorted.bam data.fasta. Before proceeding, though, I must warn you that global reconstruction is a difficult problem. Low diversity of the sample, short reads, and conserved regions, al contribute to low reliability of the results. Maybe enough to just run a local reconstruction?

Let us know.

[yorickdevries]

Have you tried to change the size of your window ("-w") to a smaller size? I solved this error by changing the window size to somewhat smaller than my average read size as suggested on http://cbg-ethz.github.io/shorah/local.html

[terrycojones]

Hi. Thanks for the replies! I'm a bit busy on a couple of other things right now, but will get back to this soon I hope.