Sam2conseq script will report "N" instead of majority consensus

[ewong347]

Expected behavior: sam2conseq should choose NT with highest frequency if the minor NT is only observed once.
Observed behavior: sam2conseq reports N instead of taking the major Nucleotide, even if the ratio is 350: 1 (position 7676)
WA11-UW7 (SRR11278092), aligned sequence, depth freqs found in langley/covid/problematic
pos 224

Freqs table:

pos	A	C	G	T	N	-	ins
224	0	0	1	15	0	0	{}
237	0	1	19	0	0	0	{}
441	0	0	40	1	0	0	{}
444	0	0	0	49	0	1	{}
447	0	1	0	49	0	0	{}
458	0	0	46	1	0	0	{}
7676	350	0	1	0	0	0	{}

[ArtPoon]

Thanks for reporting this @ewong347 - however, I can't reproduce this issue.
Here is what I am seeing:

[ewong347]

Weird- I just tried to reproduce this issue as well with no luck. When I re-ran the pipeline on the sequence, the new alignment also included more NT's on the 5' end of the sequence. Might just have been a hiccup on my end.

I think we can close this issue for now. I'll keep an eye out to see if I notice anything like this for the final couple sequences

[ArtPoon]

Well hiccups are not OK!
Can you please post your workflow so I can try to repro?

[ewong347]

I used the workflow on the main page for sam2conseq:

fasterq-dump SRRxxxx

cutadapt -q 20,20 -a CTGTCTCTTATACACATCT -o SRRxxxx.trim.fastq SRRxxxx.fastq

bowtie2 -x NC_045512 -U SRRxxxx.trim.fastq -S SRRxxxx.sam --local

python3 sam2conseq.py --unpaired SRRxxxx.sam freqs.csv SRRxxxx.conseq.txt

[ArtPoon]

Nothing unusual there. @ewong347 can you please re-run the last step a few times to see if you can reproduce this issue? Also please check your command history to see if there was something different in that run.