list index out of range

[ghost]

Hi, I'm trying to analyze sequenced reads from a virus and getting the following error:
(shorah.py -b ergebnisse.sorted.bam -f sequence.2133.fasta)

---

Traceback (most recent call last):
File "../Programme/shorah-master/shorah.py", line 142, in
keep_files=options.k, alpha=options.a)
File "/home/artur/Programme/shorah-master/dec.py", line 347, in main
r = aligned_reads.keys()[0]
IndexError: list index out of range

---

My dec.log looks like this:

---

INFO 2014-07-02 13:58:41,711 main 322 ../Programme/shorah-master/shorah.py -b ergebnisse.sorted.bam -f sequence.2133.fasta
DEBUG 2014-07-02 13:58:57,795 windows 129 Finished making windows
DEBUG 2014-07-02 13:58:57,796 windows 130 b2w returned 0
INFO 2014-07-02 13:58:57,796 parse_aligned_reads 94 Using file reads.fas of aligned reads
DEBUG 2014-07-02 13:58:57,796 parse_aligned_reads 97 Parsing aligned reads

---

I suppose it has something to do with the fact that no reads are found.
But why ? Is there a threshold to the frequency of variants?
Because the frequencies are very low.
Furthermore i wanted to specify the region for the haplotypes with the -r parameter.
Is this right ? If its right, then what is the chromosome name of a virus genome?
(-r chr:start-stop)

[ozagordi]

Hi,
it looks to me that your first problem is the same reported in issue #4, i.e. a region with few or no reads. See what the other user reported to solve it.

As for your second question, the chromosome name is simply the id of the sequence you are using as reference. Please let me know if this is fine for you. I'll close the issue in a few days.

[ghost]

Thanks for the fast answer,
but its still not working.
I resized the region for reconstruct, but all windows contain 0 reads.
I suppose its because my SNVs are too rare.
Average frequencies are 2-3*10^-4.
=> But well the Issue is not the program but my reads

[ozagordi]

Reads are assigned to windows based only on their alignment, regardless of SNVs. Check that your alignment is working by inspecting the BAM file (with samtools tview, for example).

That said, I'm not sure you will be able to reliably detect SNVs at frequency 10^-4.

[ghost]

hmm the alignment looks fine, if I get it right.
well i just don't have alignments of the reverse strand cause i do several filtering steps and realignments.
Does shorah consider the base quality?

[ozagordi]

No. What is your average read length again? And what is your window size?

[ghost]

window size is 201 didn't change it.
the read length varies from 10 to 107 but the average should be 90

[ozagordi]

The window size should be slightly smaller than the read length. I would run with 90.

[ghost]

Ah didn't read the local part.
Thanks a lot it seems to work, but there is a warning :

---

/usr/local/lib/python2.7/dist-packages/Bio/Seq.py:302: BiopythonDeprecationWarning: This method is obsolete; please use str(my_seq) instead of my_seq.tostring().
BiopythonDeprecationWarning)

---

But its probably the problem of Biopython
Thanks a lot again!