tophat and bowtie different output?

[zinky]

Hi,
today,I employed both bowtie and tophat to preform alignment of my RNA-seq data. I thought there must be a little difference between their result,however, both results were totally beyond my mind,here are my command as well as it's outputwith the same data and genome)

bowtie command:
bowtie -p 6 --chunkmbs 200 -m 2 --sam crassa_reference -1 Dis3_f_QA_L1_1.fq -2 Dis3_f_QA_L1_2.fq f_aligned.sam
output:
# reads processed: 19798076
# reads with at least one reported alignment: 35 (0.00%)
# reads that failed to align: 19798041 (100.00%)
Reported 35 paired-end alignments to 1 output stream(s)
tophat command:
tophat -p 8 -G genes.gtf -o f_thout carssa_reference Dis3_f_QA_L1_1.fq Dis3_f_QA_L1_2.fq
tophat output：
# reads processed: 19797774
# reads with at least one reported alignment: 15504241 (78.31%)
# reads that failed to align: 4293396 (21.69%)
# reads with alignments suppressed due to -m: 137 (0.00%)
Reported 15605745 alignments to 1 output stream(s)

these outputs makes me confused ,any help would be greatly appreciate

[dpryan]

You might describe what you find confusing about that. BTW, I assume that the "crassa_reference" in the first command and the "carssa_reference" in the second were originally spelled the same.

[zinky]

You might describe what you find confusing about that. BTW, I assume that the "crassa_reference" in the first command and the "carssa_reference" in the second were originally spelled the same.

thanks for your reply!
yep,it's my typing mistake, but the command in hpc is surely the same .my problem here is that the bowtie result show's 100% ummapped but tophat shows 80% mapped.

[dpryan]

I wouldn't expect bowtie to be very good at mapping spliced reads to the genome. Having said that, it is a bit surprising that you didn't get at least 20% mapping with bowtie (that's just a guesstimate on my part). You might look at a few of the Tophat alignments in IGV or SeqMonk or similar and see if you can visually see why bowtie might be having issues (presumably there's a lot of splicing or something). That will also let you know if, instead, tophat is acting oddly.

[fengqi]

Hi, All,

I have similar question, but opporsite situation.

I have RNA-Seq single end reads generate by Solid (50bp).

I also employed both bowtie and tophat to preform alignment, both using default setting.

I got less alignment reads by Tophat.

For total ~67m reads, tophat gave ~12m mapped reads, in which there are ~6m unique reads.

But bowtie give ~30m unique mapped reads.

Anyone knows what may cause that?

And another related question, when I look the mapping result in IGV, I noticed that for a same read, the tophat result gives Cigar as 50M which bowtie gives Cigar 48M. Always missing the first and the last base in bowtie.

Thanks for any thoughts

[carmeyeii]

So you can supply TopHat with a GTF file of annotated transcripts, which, using the --GTF option, will be the first place where reads are mapped, followed by the whole genome, with or without novel junction discovery in this second stage. As I understand it, this is after TopHat 1.4.
I'm curious to know how t was before 1.4. I think you could already give TopHat a GTF file, but it used it second. Am I right? If so, what is the difference between using it [the GTF file] first and using it second after the genome?