Bwa producing no alignment
0
1
Entering edit mode
7.5 years ago
ThePresident ▴ 180

I am dealing with Illumina paired-end, 150 bp read-length datasets.

Here's what my fastq files looklike:

@SRR3405394.1 1 length=151
NCGACTGAGGTAATTACAGTTCTTCGGTTCCAGCCAGCTGTCTCAGTTTATGGACCAGAACAACCCGCTGTCTGAGATTACGCACAAACGTCGTATCTCCGCACTCGGCCCAGGCGGTCTGACCCGTGAACGTGCAGGGTACAGATCGGAA
+SRR3405394.1 1 length=151
#<GGGGAGGIGGGIGGIIIIIGIIIIIIIIIIGIIIIIIIGGIIGIGGIIIIIGIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIGIGGIGGIIGGIIIIIGGGIIIIIGGIIIIGIIIGIGII<GGGGAGIGGAGG

I use this command to align with bwa:

bwa aln -t8 index 1.fastq > 1.sai
bwa aln -t8 index 2.fastq > 2.sai
bwa sampe index 1.sai 2.sai 1.fastq 2.fastq > aln_bwa.sam

When I look at my SAM file, I see this:

@SQ SN:selected LN:4641665
@PG ID:bwa  PN:bwa  VN:0.7.13-r1126 CL:bwa sampe index 1.sai 2.sai 1.fastq 2.fastq
SRR3405394.1    77  *   0   0   *   *   0   0   NCGACTGAGGTAATTACAGTTCTTCGGTTCCAGCCAGCTGTCTCAGTTTATGGACCAGAACAACCCGCTGTCTGAGATTACGCACAAACGTCGTATCTCCGCACTCGGCCCAGGCGGTCTGACCCGTGAACGTGCAGGGTACAGATCGGAA #<GGGGAGGIGGGIGGIIIIIGIIIIIIIIIIGIIIIIIIGGIIGIGGIIIIIGIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIGIGGIGGIIGGIIIIIGGGIIIIIGGIIIIGIIIGIGII<GGGGAGIGGAGG
SRR3405394.1    141 *   0   0   *   *   0   0   GTACCCTGCACGTTCACGGGTCAGACCGCCTGGGCCGAGTGCGGAGATACGACGTTTGTGCGTAATCTCAGACAGCGGGTTGTTCTGGTCCATAAACTGAGACAGCTGGCTGGAACCGAAGAACTGTAATTACCTCAGTCGTAGATCGGAA GGAAGIIGIIIGIGIIGGGIIGGGGIGGIIAGGGIIGGGGGIGIGGGGIIGGGIGGGGGIIIIIGGIIIIIIIIIIIIGIIIGIIIIGIIIGGIIIGIIIIAGGIIIGGAGIGGGGGGI<GGGGIIIIIGGGIGIIIGIIIIIAGGIIIIG

All my reads are flagged 77 / 141, i.e. they're not aligned. What's the problem? I have a feeling that the quality scores are not recognized, but that's just a wild guess based on some quick reading.

Thank you in advance, TP

bwa • 3.2k views
ADD COMMENT
2
Entering edit mode

Do you have a reason to use bwa aln rather than bwa mem?

ADD REPLY
1
Entering edit mode

I am interested in extracting aberrant pairs (same orientation fwd-fwd and bkw-bkw combinations). I know that bwa aln is able to mark them properly but I am not sure about bwa mem. Do you happen to know if bwa mem handles these type of pairs the same way as bwa aln does?

ADD REPLY
0
Entering edit mode

Just out of curiosity: why are you running aln and index at the same time? For what I remember aln is used to align and should create a sai output, while index is used to create an index of a fasta reference file.

ADD REPLY
1
Entering edit mode

I guess index is a placeholder for the path to his index file?

ADD REPLY
0
Entering edit mode

Correct, index is placeholder for my index file.

ADD REPLY
0
Entering edit mode

Is it possible that you have quality in Illumina 1.3+ or 1.5+ while (I think) bwa is now expecting Illumina 1.8+? Try giving a look at these posts:

What Is The Default Quality Encoding Expected By Bwa?

Convert Illumina 1.3 To Illumina 1.5

https://en.wikipedia.org/wiki/FASTQ_format

ADD REPLY
0
Entering edit mode

I did this awhile ago but I believe I verified that I have 1.8+ Illumina score. If I remember correctly FastQC can detect what scores are used.

ADD REPLY
0
Entering edit mode

mmm, yes, I noticed you have # in the quality, which cannot be Illumina 1.3+ nor 1.5+.

1) Why do you think it is something in the quality? Did you receive some warning?

2) Have you tried building a small reference identical to one of your reads and aligning against it?

ADD REPLY
0
Entering edit mode

1) Googling here and there led me to think that quality scores might be a problem but now I know it shouldn't be. 2) Nope, but that's a good idea. I'll try it.

ADD REPLY

Login before adding your answer.

Traffic: 2628 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6