Tophat for single end reads

[lynn012]

Hello,
I test tophat on the test data set, it runs correctly.
But when I run tophat with my single end RNA sequences, tophat_out file only contains logs and tmp, which doesn't have accepted_hits.sam, junctions.bed, insertions.bed and deletions.bed. I don't know why and where is my error.
ps: I just use default parameters.(tophat a_ref a.fq)
Thanks for your help!!!

[Camg]

Did you create an index of you reference sequence using bowtie-build?
Do you get an error message when Tophat runs? If so, what does it say?

[lynn012]

Camg, I have created an index of my reference sequence, and Tophat runs without a hitch.
The only problem is the result. Tophat_out file contains 'tmp' file besides 'logs', which includes left_kept_reads.fq，left_kept_reads_missing.fq and so on. However there isn't accepted_hits.sam.

[Camg]

Can you share the exact command that you wrote to run Tophat, as well as what Tophat reported while it was running?
What is in the left_kept_reads etc. files? I don't have those in my "logs" file.

Does anyone else have any suggestions?

[edge]

Hi, you already fix your problem?
Do you mind to share the command that you try for running single-end read in Tophat?
Thanks a lot.
I just start trying on it now as well.

[Gonza]

NO accepted_hits.bam

Hi all,

I am having a similar problem with illumina single end. Tophat is running without a problem, but I only see 'temp' and 'logs' folder (no accepted_hits.bam).

My script for one library:

tophat -p 20 -G TAIR10_GFF3_genes.gff -o 01_tophat2_TRT_R3.fq.Q30.L50 TAIR10_chr_all TRT_R3.fq.Q30.L50

I have indexed the genome using bowtie2-build. What could this be????

many thanks!

G

[GenoMax]

Go into the "logs" folder and start looking at the logs to get some additional detail. There should be an error logged in there someplace.

[Gonza]

Thanks so much GenoMax, I did find the problem within the log folder (Error: Couldn't build bowtie index with err = 1) please see below. Does this mean i have to use bowtie2-build for the genes file (TAIR10_GFF3_genes.gff)???. I downloaded both the Arabidopsis genome and genes from (ftp://ftp.arabidopsis.org/home/tair/...omosome_files/) and (ftp://ftp.arabidopsis.org/home/tair/...GFF3_genes.gff)

I have read others posts but i can't quite figure out how to solve this. I appreciate your help.


[2014-09-10 16:17:49] Beginning TopHat run (v2.0.11)
-----------------------------------------------
[2014-09-10 16:17:49] Checking for Bowtie
Bowtie version: 2.1.0.0
[2014-09-10 16:17:49] Checking for Samtools
Samtools version: 0.1.18.0
[2014-09-10 16:17:49] Checking for Bowtie index files (genome)..
[2014-09-10 16:17:49] Checking for reference FASTA file
Warning: Could not find FASTA file TAIR10_chr_all.fa
[2014-09-10 16:17:49] Reconstituting reference FASTA file from Bowtie index
Executing: /usr/local/bin/bowtie2-inspect TAIR10_chr_all > 01_tophat2_CTR_R1.fq.Q30.L50/tmp/TAIR10_chr_all.fa
[2014-09-10 16:17:56] Generating SAM header for TAIR10_chr_all
[2014-09-10 16:17:56] Reading known junctions from GTF file
[2014-09-10 16:17:59] Preparing reads
left reads: min. length=50, max. length=101, 22866275 kept reads (865 discarded)
[2014-09-10 16:30:19] Building transcriptome data files 01_tophat2_CTR_R1.fq.Q30.L50/tmp/TAIR10_GFF3_genes
[2014-09-10 16:30:22] Building Bowtie index from TAIR10_GFF3_genes.fa
[FAILED]
Error: Couldn't build bowtie index with err = 1

[GenoMax]

Instead of struggling with these individual files can you download the sequence/index/annotations package for TAIR 10 from the iGenomes site here: http://support.illumina.com/sequenci...e/igenome.html

[Gonza]

Thanks, will try this. Does it matter if you download TAIR10 from Ensembl or NCBI?

-G

[GenoMax]

Sequence should be the same. Annotations will be different. NCBI may be better place to start since the Ensembl may have computationally predicted entries that may complicate things.

[Gonza]

Thanks so much, it worked (i see the accepted_hits.bam!!!!!!).
I had actually started the analysis with the Ensembl genome, should i downloaded the NCBI and start over?

Again, many thanks.
G

[GenoMax]

No need to start over. Just stick with Ensembl dataset (sequence/annotation) for the rest of the analysis.

[Gonza]

great. Thank you kindly for all the help.
-G

[Gonza]

Dear All,

Quick question, should you include the " -- phred64-quals" in the tophat script? I think by default tophat does it but I am not quite sure honestly.
Also, including -- phred64-quals helps to map only reads with better quality?

Thanks
G