Align.seqs and accnos

Commands in mothur
1

Dear friends of mothur,

I have a question concerning align.seqs. I have run it on a fasta file (previously “dereplicated” with unique.seqs) but I have not provided the associated count file as an argument for align.seqs (in the wiki, count is not listed as an argument for align.seqs).

I get a [WARMING] about 633 sequences whose alignment eliminated too many bases (supposedly listed in the accnos file) and a [NOTE] about 437 sequences that were flipped to produce a better alignment. The aligned fasta has the same number of lines as the unaligned fasta (no sequence was deleted).

However, the number of lines in the accnos file is the reported number of flipped sequences (not the number of bad alignments).

So my questions are: should I delete the 633 sequences that generate the [WARNING]? And if yes, how should I do this?

Yours,
Maxime

2

Hi Maxime,

You will do this with screen.seqs when you set the start and end positions.

Pat

3

Thank you Pat!

So am I right in my understanding that:

-the sequences that trigger the [WARNING] are not deleted at the align.seqs step (but they deserve to, though at a later step by screen.seqs);
-they are not the ones listed in accnos file;
-the ones in the accnos are the flipped ones, that do not deserve being deleted just on this ground?

Yours,
Maxime

4

I’m pretty sure the last two points are correct. There’s really no need to do anything wiht that accnos file

Pat

5

Thank you Pat! I will try to add this info to the wiki page on align.seqs when I have time.

Maxime

6

This topic was automatically closed 10 days after the last reply. New replies are no longer allowed.