TopHat2 on multiple samples, avoid building Bowtie index from genes.fa each time?

[LeonDK]

Hi all,

Ultimately aiming at differential expression, I'm mapping human RNAseq read using tophat2 with the following command:

tophat2 --num-threads 12 --GTF /Homo_sapiens/UCSC/hg19/Annotation/Genes/genes.gtf /Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome myfastq_R1.fastq.gz myfastq_R2.fastq.gz

Is it really necessary to:

[2014-09-18 10:38:45] Building transcriptome data files /tmp/genes
[2014-09-18 10:39:21] Building Bowtie index from genes.fa

Foreach sample? I mean - The different samples are all mapped using the same Bowtie2Index/genome files and the same Genes/genes.gtf files?

Cheers,
Leon

[dpryan]

Have a look at the --transcriptome-index option, which is what you're looking for.

[LeonDK]

Have a look at the --transcriptome-index option, which is what you're looking for.

Hi dpryan,

Thanks for input reg. the --transcriptome-index option for tophat2. I looked it up in the TopHat2 manual. For other users, which may encounter the same challenge - The trick is to run this command first:

tophat2 -G iGenomes/Homo_sapiens/UCSC/hg19/Annotation/Genes/genes.gtf --transcriptome-index=transcriptome_data/known iGenomes/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome

and then subsequently call tophat2 with this command:

tophat2 --num-threads 12 --transcriptome-index=transcriptome_data/known iGenomes/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome myfastq_R1.fastq.gz myfastq_R2.fastq.gz

After running the above command, you'll see

[2014-09-18 12:12:04] Using pre-built transcriptome data..

Which is significantly faster, when running multiple samples.

The UCSC/hg19 data can retrieved like so:

wget ftp://igenome:[email protected]/Homo_sapiens/UCSC/hg19/Homo_sapiens_UCSC_hg19.tar.gz

Cheers,
Leon

[konika]

tophat not creating transcriptome indexes

Hi
In my case The following command doesnt start tophat2. tophat2 just shows me the available options, like I have used a wrong option somewhere. Does anyone has an idea whats wrong here
The command I use:
tophat2 -G /home/chawla/rna_seq_pipeline/gff/mouse_ensembl.gff --transcriptome-index=tdata /home/chawla/rna_seq_pipeline/gff/mouse_ensembl

[GenoMax]

Hi
In my case The following command doesnt start tophat2. tophat2 just shows me the available options, like I have used a wrong option somewhere. Does anyone has an idea whats wrong here
The command I use:
tophat2 -G /home/chawla/rna_seq_pipeline/gff/mouse_ensembl.gff --transcriptome-index=tdata /home/chawla/rna_seq_pipeline/gff/mouse_ensembl

You have to point tophat2 process to the indexes for the full genome. It appears that you are including a gff file instead of the bowtie2 indexes at the end of your command. Refer to LeonDK's example in posts above.

[konika]

Thanks, it was actually old version of tophat that also needs an input read to create transcriptome indexes.