This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How do I run Tophat?

I am using the following command:

./tophat2 -r 210 -p 8 -G ~/Data/*.gtf ~/Data/hg19f/hg19 ~/Data/thyroid1,~/Data/testes1 ~/Data/thyroid2,~/Data/testes2

but I'm getting the error:

[2015-04-11 21:58:47] Checking for reference FASTA file
Warning: Could not find FASTA file /home/anoronh4/HW3/hg19f/hg19.fa
[2015-04-11 21:58:47] Reconstituting reference FASTA file from Bowtie index
Error locating program: bowtie-inspect

The files in hg19f are:

hg19.1.ebwt
hg19.3.ebwt
hg19.rev.1.ebwt
make_hg19.sh
hg19.2.ebwt
hg19.4.ebwt
hg19.rev.2.ebwt

Since when do I need a .fa file?

tophat bowtie

one correction:

./tophat2 -r 210 -p 8 -G ~/Data/*.gtf ~/Data/hg19f/hg19 ~/Data/thyroid1,~/Data/testes1 ~/Data/thyroid2,~/Data/testes2

Give full names of fasta/q files. Why its *.gtf ?

Is bowtie2 installed? Is bowtie-inspect in the path?

3 answers

Hi,

I think you want to align RNAseq reads to your reference genomes. If so, and If you have paired-end reads of RNAseq (R1 and R2), you should follow these steps:

  1. Build index files using bowtie2 (bowtie2 is for tophat2) using bowtie2-build your_fasta_file. After generating bowtie index files, do not compromise them, just leave in your directory. In your case "The files in hg19f are:" you put all bowtie index files in a file and used the file in your tophat command. you should not do this. first, you will need to use your fast file to make bowtie index file, you will use your the same genome file in tophat command, not index files.
  2. Run tophat command. in tophat command, you need to use your fasta (genome file) and RNAseq files (R1 and R2). The fast file is in .fa format. Ransom files are generally in .gz (gunzip) format.

Here is a sample tophat command:

tophat \
  -o OUT.dir2 \
  --library-type fr-secondstrand \
  --mate-std-dev 50 \
  -a 6 \
  --microexon-search \
  --min-segment-intron 10 \
  --max-segment-intron 20000 \
  -i 30 \
  -I 20000 \
  -r 0 \
  genome.fa \
  R1.fastq.gz R2.fastq.gz

Hello sir I followed the instructions that you told about RNAseq but I'm getting an error like this:

Error: Could not find Bowtie 2 index files (hg38.fa.*.bt2l).

Hi , Sorry for the delay. Before running Tophat, you have to make an index file that will be used by Tophat.

For this;

bowtie2-build your fasta file indexname

just give any index name (for example: myindexname).

Then, in Tophat analysis, for the argument index in Tophat command, just type myindexname and run.

Hello I run the Tophat command as you mention in the last post. The program run but throw me a message at the end:

[2017-04-05 09:53:57] Checking for reference FASTA file
Warning: Could not find FASTA file TAIR10_chr_all.fa
[2017-04-05 09:53:57] Reconstituting reference FASTA file from Bowtie index
  Executing: /data/software/bowtie2-2.2.9/bowtie2-inspect TAIR10_chr_all > tophat_mapping/tmp/TAIR10_chr_all.fa

Is that message normal?

Very good protocol for learning tuxedo! However RNA-STAR is way faster than tophat and bowtie.

It was because I had added some but not all bowtie functionalities to my path. A silly oversight.

Log in to answer this question.