Filter.seqs error: Sequences are not all the same length

1

Hi Mothur,

I am trying to analyze samples from v3-v4 region following the SOP, I have alligned the sequences and screened the alligned fasta. Then comes the filter.seqs step. Here is the sumary.seqs of my input to filter.seqs and the filter.seqs command as I am using it.

mothur > summary.seqs(fasta=../output_files/stability.trim.contigs.good.unique.good.align)

Using 128 processors.

                Start   End     NBases  Ambigs  Polymer NumSeqs
Minimum:        1       3983    165     0       3       1
2.5%-tile:      1       18929   440     0       4       33561
25%-tile:       1       18929   459     0       5       335602
Median:         1       18929   460     0       5       671203
75%-tile:       1       18929   465     0       6       1006804
97.5%-tile:     1       18929   465     0       6       1308844
Maximum:        1       18929   470     0       8       1342404
Mean:   1       18928   457     0       5
# of Seqs:      1342404

It took 11 secs to summarize 1342404 sequences.

Output File Names:
/home/jup71eb/Project_Ceyda/output_files/stability.trim.contigs.good.unique.good.summary


mothur > filter.seqs(fasta=../output_files/stability.trim.contigs.good.unique.good.align, vertical=T, trump=.)

and after a few seconds I get the error:

[ERROR]: Sequences are not all the same length, please correct.
10456
[ERROR]: Sequences are not all the same length, please correct.
10405
10408
10408
10347
10429
Segmentation fault (core dumped)

Thanks
Juan

2

Hi Juan,

That looks weird. Could you try something before running filter.seqs? Try (re-)running screen.seqs with start=1, end=18929. Then try running filter.seqs again and see if the issue goes away.

Pat

3

Dear Mothur Team,

Thanks that solved the problem.

Best
Juan