bowtie2 alignment has 'Duplicated sequence'

Question

[rmash]

I built an index using the following fasta file:

INDEX FASTA (example):

>miRNA:mmu-mir-23b MI0000141 Mus musculus miR-23b stem-loop
GGCTGCTTGGGTTCCTGGCATGCTGATTTGTGACTTGAGATTAAAATCACATTGCCAGGG
ATTACCACGCAACC
>miRNA:mmu-mir-27b MI0000142 Mus musculus miR-27b stem-loop
AGGTGCAGAGCTTAGCTGATTGGTGAACAGTGATTGGTTTCCGCTTTGTTCACAGTGGCT
AAGTTCTGCACCT

I then took my FASTQ data (example):

@K00252:57:HGFMMBBXX:1:1101:4320:1209 1:N:0:AGTCAA
NCCCTGTAGATCCGAATTTGTG
+
#AAFFJJJJJJJJJJJJJJJJJ
@K00252:57:HGFMMBBXX:1:1101:5132:1209 1:N:0:AGTCAA
NAACGGAATCCCAAAAGCAGCTG
+
#AAFFJJJJJJJJJJJJJJJJJJ

Used bowtie2 to align but ended up with a bizarre sam file that said I had duplicated entires

OUTPUT SAM looks like this:

@HD     VN:1.0  SO:unsorted
@SQ     SN:CONTAMINATION:ADAPTER:adapters_contam1       LN:100
@SQ     SN:miRNA:mmu-let-7g     LN:88
@SQ     SN:miRNA:mmu-let-7i     LN:85
@SQ     SN:miRNA:mmu-mir-1a-1   LN:77
@SQ     SN:miRNA:mmu-mir-15b    LN:64
@SQ     SN:miRNA:mmu-mir-23b    LN:74
@SQ     SN:miRNA:mmu-mir-27b    LN:73
@SQ     SN:miRNA:mmu-mir-29b-1  LN:71
@SQ     SN:miRNA:mmu-mir-30a    LN:71
@SQ     SN:miRNA:mmu-mir-30b    LN:96

Any ideas what I might be doing wrong?

[rmash]

I wonder if this has to do with mapping small RNAs which are likely to map to multiple regions?

[GenoMax]

You should use bowtie v.1 if you are mapping small RNA's where you want to do ungapped alignments.

Answer 1

[h.mon]

What you are seeing are the sam headers, not alignments. Looks normal top me.

that said I had duplicated entires

How exactly it said it? Do you mean bowtie summary at the end of alignment?

[rmash]

Doesn't seem to have any of other other tags (e.g. HI, and I can't seem to sort the sam file)

[h.mon]

Show the commands used and the error messages.

[rmash]

I am running a script collapse.py (input output) which has worked before

rmash$ python collapse.py filename.sam filename.fasta
[W::sam_hdr_parse] Duplicated sequence 'rRNA:12S_rRNA'
Traceback (most recent call last):
  File "collapse.py", line 25, in <module>
    entries = [entry for entry in pysam.AlignmentFile(args["INFNAME"],"rb") if dict(entry.get_tags())["HI"] == 1]# only get the first entry in case of multi-mappers
KeyError: 'HI'

[h.mon]

Instead of going back and forth for a long time, why don't you describe everything you did - which programs (possibly with links to their sites), command lines, reference genome used, and so on.

I believe collapse.py (which I don't know from which pipeline is, as there are several collapse.py around) is complaing about duplicated sequences on the reference, not on your reads. Check with:

grep "rRNA:12S_rRNA" filename.fasta

And:

samtools view filename.fasta | grep "rRNA:12S_rRNA"

[rmash]

like this:

[W::sam_hdr_parse] Duplicated sequence 'rRNA:12S_rRNA'

and when you try to sort it lists lots as being duplicated