why can't I get alignment with BWA?

[gigigou]

What's the general steps to align a set of reads to ref?
Is the flowing right?
1.build index -> bwa index -c -a is ecoli.fa
2.convert csfasta and correspond qual file into fastq
3.Find the SA coordinates of the input reads -> bwa aln -n 3 -t 4 -M 3 -c ecoli.fa t_ecoli.fastq >map.sai
4.Generate alignments in the SAM format given single-end reads -> bwa samse -n 1 ecoli.fa map.sai t_ecoli.fastq >map.sam
Do as the above, I get a sam file with no reads mapped. All the 25000 items have flag 4.
What's more, I have 50000 reads in the input fastq file, why does it only process 25000?
If I delete one read form the fastq file, BWA will process 24999 reads. Why?
And what is the bwasw for?
Thank you !

[jimmybee]

I think in step 3 you are missing the output character. It should be:

bwa aln -n 3 -t 4 -M 3 -c ecoli.fa t_ecoli.fastq > map.sai

See if that helps

[gigigou]

I think in step 3 you are missing the output character. It should be:

bwa aln -n 3 -t 4 -M 3 -c ecoli.fa t_ecoli.fastq > map.sai

See if that helps

Oh, sorry that I missed the ">" in the post, but I did add it in the command line.
So it can't be the reason.
Actually I do get the align file, but in the file I find there is no mapped items. All the 25000 reads are reported unmapped.
thank you all the same

[swbarnes2]

What's the general steps to align a set of reads to ref?
Is the flowing right?
1.build index -> bwa index -c -a is ecoli.fa
2.convert csfasta and correspond qual file into fastq
3.Find the SA coordinates of the input reads -> bwa aln -n 3 -t 4 -M 3 -c ecoli.fa t_ecoli.fastq >map.sai
4.Generate alignments in the SAM format given single-end reads -> bwa samse -n 1 ecoli.fa map.sai t_ecoli.fastq >map.sam
Do as the above, I get a sam file with no reads mapped. All the 25000 items have flag 4.
What's more, I have 50000 reads in the input fastq file, why does it only process 25000?
If I delete one read form the fastq file, BWA will process 24999 reads. Why?
And what is the bwasw for?
Thank you !

You are making a color space index, but converting your reads to fastq? Are you sure that's right?

[westerman]

I agree with swbarnes, after creating the color-space index then give bwa the color-space input file and not a nucleotide file.

[gigigou]

You are making a color space index, but converting your reads to fastq? Are you sure that's right?

The reads are converted into fastq format, but they are still in color space
@SRR001354.lite.sra.1 461_28_1048
T23331323333132332323133333332333232
+
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

I tried to input the original csfasta file, it processed 50000 reads, but still with no reads mapped.

50000 + 0 in total (QC-passed reads + QC-failed reads)
0 + 0 duplicates
0 + 0 mapped (0.00%:-nan%)
0 + 0 paired in sequencing
0 + 0 read1
0 + 0 read2
0 + 0 properly paired (-nan%:-nan%)
0 + 0 with itself and mate mapped
0 + 0 singletons (-nan%:-nan%)
0 + 0 with mate mapped to a different chr
0 + 0 with mate mapped to a different chr (mapQ>=5)

[gigigou]

I agree with swbarnes, after creating the color-space index then give bwa the color-space input file and not a nucleotide file.

In the fsatq file that converted from csfasta, the reads are still in color space, as above.

[westerman]

Since we do not have access to your files it is hard to troubleshoot problems. But a couple of pieces of general advice in troubleshooting a problem:

1) Do not use any non-standard options; e.g., try 'aln' and 'samse' without any parameters.

2) Try a smaller input file. Especially one with a known-to-be-good sequence. In other words take a bit of your ecoli reference and make it into a color-space read.

Good luck in solving this.

[GenoMax]

Provided you have done what others have already suggested before ...

(and this may be a stupid suggestion) but are you sure the data you have is from your samples (i.e. there was no mix-up at the place where got it sequenced).

[gigigou]

Since we do not have access to your files it is hard to troubleshoot problems. But a couple of pieces of general advice in troubleshooting a problem:

1) Do not use any non-standard options; e.g., try 'aln' and 'samse' without any parameters.

2) Try a smaller input file. Especially one with a known-to-be-good sequence. In other words take a bit of your ecoli reference and make it into a color-space read.

Good luck in solving this.

Thank you.
All the parameters are specified as the manual says.
I think the input file is small enough, only 50000 reads.
Thank you for your help.
I'll look into it.

[gigigou]

Provided you have done what others have already suggested before ...

(and this may be a stupid suggestion) but are you sure the data you have is from your samples (i.e. there was no mix-up at the place where got it sequenced).

I think the data is not the problem cause I have used other align tools to do the alignment, and they all give perfect results. So I think the problem is BWA, maybe I didn't specify the parameters properly, I'll try to solve it.
Thank you!

[gigigou]

I have solved the problem.
In step 2, the file's name should be specified as the pl says.
But I think it is a little inconvenient, so I modified the pl to let it work more efficiently.