wrong bam file format

Question

[alirezamomeni707]

I am trying to align my .fastq file using tophat on transcriptom. the resulting .bam file look like this:

@SQ SN:89720    LN:1312
@SQ SN:89721    LN:735
@SQ SN:89722    LN:5191
@SQ SN:89723    LN:361
@SQ SN:89724    LN:9056
@SQ SN:89725    LN:83
@SQ SN:89726    LN:2603
@SQ SN:89727    LN:954
@SQ SN:89728    LN:468

but when I try to make .bg or .wig file for the UCSC, I need to have some thing like this:

@SQ SN:chr1 LN:249250621
@SQ SN:chr10    LN:135534747
@SQ SN:chr11    LN:135006516
@SQ SN:chr12    LN:133851895
@SQ SN:chr13    LN:115169878
@SQ SN:chr14    LN:107349540
@SQ SN:chr15    LN:102531392
@SQ SN:chr16    LN:90354753
@SQ SN:chr17    LN:81195210

otherwise UCSC gives error (because the chromosomes are not clear in the first one). do you know what the problem is?

Answer 1

[Devon Ryan]

Delete the BAM file, do not use tophat, and realign the data. Common aligners for RNAseq data are STAR, hisat2, and BBMap.

Answer 2

[Ido Tamir]

Because you aligned to the transcriptome. With tophat you align to the genome + a transcriptome annotation in GTF format if you have one at hand. And check the names of the chromosomes. ENSEMBL does not do the chr prefix. You might have to change it in the files.

[alirezamomeni707]

no because I had a look at the reference file I used and chr numbers are there.

[GenoMax]

It is not just numbers. 17 is not equivalent to chr17. In your case the chromosome names seem to be some odd numbers (e.g. 89720). Are these coming from contig headers/fasta files?

[Ido Tamir]

do grep '>' genome.fa Also: the headers you have now look like from transcripts. The second column is the length of the contig. If you have additional chromosomes, UCSC does not care. It only cares if it cant find the one it wants to display in your bam file. Please write the UCSC error into the question. Maybe you even have a different problem.