Tophat2 returns empty junction file: Warning: junction database is empty!

[sachitad]

I am using Tophat2 V2.0.7 with this command:

tophat2 -p 1 --library-type fr-unstranded --segment-length 12 /data/rathi/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome Galaxy164-\[FASTQ_Groomer_on_data_152\].fastqsanger

#Running:

[2013-02-14 03:31:11] Beginning TopHat run (v2.0.7)
-----------------------------------------------
[2013-02-14 03:31:11] Checking for Bowtie
		  Bowtie version:	 2.0.6.0
[2013-02-14 03:31:11] Checking for Samtools
		Samtools version:	 0.1.18.0
[2013-02-14 03:31:11] Checking for Bowtie index files
[2013-02-14 03:31:11] Checking for reference FASTA file
	Warning: Could not find FASTA file /data/rathi/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome.fa
[2013-02-14 03:31:11] Reconstituting reference FASTA file from Bowtie index
  Executing: /usr/bin/bowtie2-inspect /data/rathi/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome > ./tophat_out/tmp/genome.fa
[2013-02-14 03:33:21] Generating SAM header for /data/rathi/Homo_sapiens/UCSC/hg19/Sequence/Bowtie2Index/genome
	format:		 fastq
	quality scale:	 phred33 (default)
[2013-02-14 03:33:54] Preparing reads
	 left reads: min. length=36, max. length=36, 2 kept reads (0 discarded)
[2013-02-14 03:33:54] Mapping left_kept_reads to genome genome with Bowtie2 
[2013-02-14 03:34:26] Mapping left_kept_reads_seg1 to genome genome with Bowtie2 (1/3)
[2013-02-14 03:34:59] Mapping left_kept_reads_seg2 to genome genome with Bowtie2 (2/3)
[2013-02-14 03:35:32] Mapping left_kept_reads_seg3 to genome genome with Bowtie2 (3/3)
[2013-02-14 03:36:05] Searching for junctions via segment mapping
Warning: junction database is empty!
[2013-02-14 03:37:27] Reporting output tracks
-----------------------------------------------
[2013-02-14 03:38:48] Run complete: 00:07:37 elapsed

The only thing I get ouput in junctions.bed is:

track name=junctions description="TopHat junctions"

What am I doing wrong?

[iris_aurelia]

It seems that you only have 2 reads in your dataset (2 kept reads). I would try a bigger set of reads to start with.

[sachitad]

I tried with all the possible data but am getting the same error about junction. Is it because of the reference genome?

[iris_aurelia]

Have you checked if the created reference fasta file wasn't empty? Mayby you could try to recreate the reference file upon analysis using bowtie2-inspect?

[sachitad]

I already have .fa file in Bowtie2Index directory and it's not empty:

These are files with reference gnome:

genome.1.bt2 genome.2.bt2 genome.4.bt2 genome.fa.fai genome.rev.2.bt2 reads_2.fq
genome.3.bt2 genome.fa genome.rev.1.bt2 reads_1.fq