align.seqs command killed after sometime

Commands in mothur
1

Hi Everyone,
I have a 16s DNA metagenomic data and I am using align.seqs command to align the reads to ribosomal database. I have downloaded aligned sequences from the ribosomal database. When I am running the command after sometime it get killed, I think because it is using lot of memory. I do not know how to solve this issue, please anyone can help me in this.
Thanks for any help!!!

2

Can you post the command text you are using? How much RAM does your computer have?

3

Thanks for looking into my problem, the command I am using is
align.seqs(fasta=…/demultiplexed.fna, reference=…/rdp.fa, flip=T,search=blast, align=blast)
I am doing this on server which has 32Gb of memory.
The rdp database is around 68 Gb, this is the aligned database I am using, before I used unaligned database and it was fine but when I was using filter.seqs command after aligning the reads. It says that the sequence are of unequal length.
Can you suggest any other database for 16s DNA alignment.
Thanks!!1

4

So you’re saying that rdp.fa is not aligned? That’s not going to work…

5

When I was using unaligned database then it was going fine, but now I am using aligned database and it is getting killed every time.
The database is very big itself.

Thanks!!!

6

rdp.fa is aligned

7

If I had to guess, 68GB won’t fit into 32GB of RAM. Why not use our silva.bacteria.fasta reference? It’s a lot better than the RDP…

8

Thanks for the reply Schloss,
I used greengenes database for alignment, but when I am using filter.seq command then it is giving me error that sequence are not of the same length.
is that database not aligned.
I downloaded it from here
http://www.mothur.org/wiki/Schloss_SOP
Thanks!!!

9

Not sure what you were using… Use this:

http://www.mothur.org/w/images/9/98/Silva.bacteria.zip

10

Thanks Pat,
I am now using this.

11

Hi Pat,
I used Silva database as suggested by you, my reads got aligned and then I did the next step which is filtering the sequences by using command filter.seqs(fasta=…/demultiplexed.u.align, vertical=T, trump=., processors=2) When I am running this command I am getting this error can you tell me why I am getting this error. I thought when the databases are not aligned we get this type of errors.
The error is:
Creating Filter…
Sequences are not all the same length, please correct.
Sequences are not all the same length, please correct.
My summary file looks like this:
Start End NBases Ambigs Polymer NumSeqs
Minimum: 0 0 0 0 1 1
2.5%-tile: -1 -1 0 0 1 194
25%-tile: 13862 21777 151 0 4 1933
Median: 13862 21777 151 0 4 3865
75%-tile: 13862 21777 151 0 5 5797
97.5%-tile: 13870 21780 151 0 6 7535
Maximum: 15883 22438 151 0 8 7728
Mean: 13451.4 21117.5 146.075 0 4.28429

of Seqs: 7728


Please help!!!
12

can you provide the exact commands and ordering you are using (and version of mothur you are using)?

13

Hi Pat,
The command I used is
filter.seqs(fasta=…/demultiplexed.u.align, vertical=T, trump=., processors=2)

I did not screened the sequences using screen command. But, once I did screen.seqs I do not get the error:
The screen command is
screen.seqs(fasta=…/demultiplexed.u.align, start=13862, end=21777, maxhomop=6)

I hope I am going on the right path.
Also, when I aligned my sequences with archea I get this summary file :
Start End NBases Ambigs Polymer NumSeqs
Minimum: 0 0 0 0 1 1
2.5%-tile: -1 -1 0 0 1 194
25%-tile: -1 -1 0 0 1 1933
Median: -1 -1 0 0 1 3865
75%-tile: -1 -1 0 0 1 5797
97.5%-tile: -1 -1 0 0 1 7535
Maximum: 14974 21780 151 0 5 7728
Mean: 81.8675 128.375 0.876423 0 1.0198

of Seqs: 7728

Does this mean that none of my sequences got hit in archae.
The mothur version I am using is mothur v.1.27.0

Thanks!!!