puzzle with gmapper -M mirna (shrimp)

Sergio.pv · 11-21-2013, 05:55 AM

Hi everyone,
i am mapping short reads from ion torrent PGM. Reads are QV~20 and 16-155 length.
In FastQC appears this as the highest represented read:
TGAGGTAGTAGGTTGTGTGGTT 19528counts

Runing a quick blast it matches 100% bta-let-7b (bta my sp of interest):
UserSe 1 ugagguaguagguugugugguu 22
bta-let-7b 1 ugagguaguagguugugugguu 22

Of course I can't do this with all my reads, so i mapped them to all ncRNAs and mirna-stem-loops for bos taurus, using shrimp in mirna mode:

gmapper-ls myfile.fa P1_adapter.fa bta-ncRNAs.fa -M mirna >myfile.out 2>myfile.log

After counting the repeated elements in the RNAME coulumn of the SAM file, bta-let-7b appears nowhere and neither the corresponding miRNAs for the other highly counted reads.

I have tried this again using same parameters with only stem loops and then only mature mirna sequences.

In all trials the resuls were the same.

Hope you can come up with soemthing, i will keep trying other settings for gmapper and I will let you know how it went.

Thaks!

Sergio.pv · 11-21-2013, 07:16 AM

I did the following, as a test:
first, i use the above metioned sequences (identical as you can see)
>bta-let-7b MIMAT0004331
UGAGGUAGUAGGUUGUGUGGUU
>dummy
UGAGGUAGUAGGUUGUGUGGUU

gmapper-ls dummy-seq.fa dummy-let-7b.fa -N 12 -o 10 -h 80% >map.out 2>map.log

Output:
HD VN:1.0 SO:unsorted
@SQ SN:bta-let-7b LN:22
@PG ID:gmapper VN:2.2.3 CL:gmapper-ls dummy-seq.fa dummy-let-7b.fa -N 12 -o 10 -h 80%
dummy 0 bta-let-7b 1 250 22M * 0 0 NGAGGNAGNAGGNNGNGNGGNN *

Which is ok.

But if I change the dummy sequence (U -> T), into the original output of the sequencer (Us as Ts) i don't get results:
@HD VN:1.0 SO:unsorted
@SQ SN:bta-let-7b LN:22
@PG ID:gmapper VN:2.2.3 CL:gmapper-ls dummy-seq.fa dummy-let-7b.fa -N 12 -o 10 -h 80%

It seems that the use of Ts is not computed. Is that correct?, is there an option to modifify that?

Best!

Sergio.pv · 11-22-2013, 02:45 AM

Ok, to make it work i converted U into T and it worked.
if it is any use, with fastx tool kit:

fasta_nucleotide_changer -v -i bta-stem-loops.mirbase20.fa -o bta-stemloops.fa -d

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
SHRiMP alignment results for colorspace miRNA	dsidote	Bioinformatics	2	11-10-2011 10:32 AM
Plasma DNA Puzzle	whw	Sample Prep / Library Generation	9	08-03-2011 11:41 AM
puzzle in fq_all2std.pl	biocc	Bioinformatics	0	11-25-2010 04:33 AM
blast puzzle	anyone1985	Bioinformatics	1	09-06-2009 02:45 AM
format puzzle	anyone1985	Illumina/Solexa	0	03-13-2009 04:29 AM

11-21-2013, 05:55 AM	#1
Sergio.pv Member Location: Berlin Join Date: Jul 2013 Posts: 20	puzzle with gmapper -M mirna (shrimp) Hi everyone, i am mapping short reads from ion torrent PGM. Reads are QV~20 and 16-155 length. In FastQC appears this as the highest represented read: TGAGGTAGTAGGTTGTGTGGTT 19528counts Runing a quick blast it matches 100% bta-let-7b (bta my sp of interest): UserSe 1 ugagguaguagguugugugguu 22 bta-let-7b 1 ugagguaguagguugugugguu 22 Of course I can't do this with all my reads, so i mapped them to all ncRNAs and mirna-stem-loops for bos taurus, using shrimp in mirna mode: gmapper-ls myfile.fa P1_adapter.fa bta-ncRNAs.fa -M mirna >myfile.out 2>myfile.log After counting the repeated elements in the RNAME coulumn of the SAM file, bta-let-7b appears nowhere and neither the corresponding miRNAs for the other highly counted reads. I have tried this again using same parameters with only stem loops and then only mature mirna sequences. In all trials the resuls were the same. Hope you can come up with soemthing, i will keep trying other settings for gmapper and I will let you know how it went. Thaks! Last edited by Sergio.pv; 11-21-2013 at 05:57 AM.

11-21-2013, 07:16 AM	#2
Sergio.pv Member Location: Berlin Join Date: Jul 2013 Posts: 20	I did the following, as a test: first, i use the above metioned sequences (identical as you can see) >bta-let-7b MIMAT0004331 UGAGGUAGUAGGUUGUGUGGUU >dummy UGAGGUAGUAGGUUGUGUGGUU gmapper-ls dummy-seq.fa dummy-let-7b.fa -N 12 -o 10 -h 80% >map.out 2>map.log Output: HD VN:1.0 SO:unsorted @SQ SN:bta-let-7b LN:22 @PG ID:gmapper VN:2.2.3 CL:gmapper-ls dummy-seq.fa dummy-let-7b.fa -N 12 -o 10 -h 80% dummy 0 bta-let-7b 1 250 22M * 0 0 NGAGGNAGNAGGNNGNGNGGNN * Which is ok. But if I change the dummy sequence (U -> T), into the original output of the sequencer (Us as Ts) i don't get results: @HD VN:1.0 SO:unsorted @SQ SN:bta-let-7b LN:22 @PG ID:gmapper VN:2.2.3 CL:gmapper-ls dummy-seq.fa dummy-let-7b.fa -N 12 -o 10 -h 80% It seems that the use of Ts is not computed. Is that correct?, is there an option to modifify that? Best!

11-22-2013, 02:45 AM	#3
Sergio.pv Member Location: Berlin Join Date: Jul 2013 Posts: 20	Ok, to make it work i converted U into T and it worked. if it is any use, with fastx tool kit: fasta_nucleotide_changer -v -i bta-stem-loops.mirbase20.fa -o bta-stemloops.fa -d