Hi,
I have a question about using TopHat. I want to map the reads only using known gene annotations (without discovering any novel splice junctions). Therefore I am using --GTF and --transcriptome-index options. However, I am not sure which option should I use --no-novel-juncs or --transcriptome-only and what is the difference between the two?
For me it seems that whichever options I choose, TopHat still tries to map reads to genome (if they do not map to transcriptome). See below the log file from a run with --transcriptome-only option.
------------
Mon Feb 6 22:18:11 2012] Preparing output location input/GM06985/tophat/
[Mon Feb 6 22:18:11 2012] Checking for Bowtie index files
[Mon Feb 6 22:18:11 2012] Checking for Bowtie index files
[Mon Feb 6 22:18:11 2012] Checking for reference FASTA file
[Mon Feb 6 22:18:11 2012] Checking for Bowtie
Bowtie version: 0.12.7.0
[Mon Feb 6 22:18:11 2012] Checking for Samtools
Samtools Version: 0.1.16
[Mon Feb 6 22:18:11 2012] Generating SAM header for hg19
format: fastq
quality scale: phred33 (default)
[Mon Feb 6 22:18:35 2012] Reading known junctions from GTF file
[Mon Feb 6 22:18:52 2012] Preparing reads
left reads: min. length=50, count=38337749
[Mon Feb 6 22:29:55 2012] Using pre-built transcriptome index..
[Mon Feb 6 22:29:55 2012] Mapping left_kept_reads against transcriptome known with Bowtie
[Mon Feb 6 23:19:26 2012] Converting left_kept_reads.m2g to genomic coordinates (map2gtf)
[Mon Feb 6 23:54:00 2012] Resuming TopHat pipeline with unmapped reads
[Mon Feb 6 23:54:00 2012] Mapping left_kept_reads.m2g_um against hg19 with Bowtie
[Tue Feb 7 01:08:53 2012] Processing bowtie hits
[Tue Feb 7 01:25:39 2012] Mapping left_kept_reads.m2g_um_seg1 against hg19 with Bowtie (1/2)
[Tue Feb 7 02:20:36 2012] Mapping left_kept_reads.m2g_um_seg2 against hg19 with Bowtie (2/2)
[Tue Feb 7 03:06:23 2012] Searching for junctions via segment mapping
[Tue Feb 7 04:43:56 2012] Retrieving sequences for splices
[Tue Feb 7 05:08:27 2012] Indexing splices
[Tue Feb 7 05:19:54 2012] Mapping left_kept_reads.m2g_um_seg1 against segment_juncs with Bowtie (1/2)
[Tue Feb 7 05:49:43 2012] Mapping left_kept_reads.m2g_um_seg2 against segment_juncs with Bowtie (2/2)
[Tue Feb 7 06:17:11 2012] Joining segment hits
[Tue Feb 7 06:36:14 2012] Reporting output tracks
I have a question about using TopHat. I want to map the reads only using known gene annotations (without discovering any novel splice junctions). Therefore I am using --GTF and --transcriptome-index options. However, I am not sure which option should I use --no-novel-juncs or --transcriptome-only and what is the difference between the two?
For me it seems that whichever options I choose, TopHat still tries to map reads to genome (if they do not map to transcriptome). See below the log file from a run with --transcriptome-only option.
------------
Mon Feb 6 22:18:11 2012] Preparing output location input/GM06985/tophat/
[Mon Feb 6 22:18:11 2012] Checking for Bowtie index files
[Mon Feb 6 22:18:11 2012] Checking for Bowtie index files
[Mon Feb 6 22:18:11 2012] Checking for reference FASTA file
[Mon Feb 6 22:18:11 2012] Checking for Bowtie
Bowtie version: 0.12.7.0
[Mon Feb 6 22:18:11 2012] Checking for Samtools
Samtools Version: 0.1.16
[Mon Feb 6 22:18:11 2012] Generating SAM header for hg19
format: fastq
quality scale: phred33 (default)
[Mon Feb 6 22:18:35 2012] Reading known junctions from GTF file
[Mon Feb 6 22:18:52 2012] Preparing reads
left reads: min. length=50, count=38337749
[Mon Feb 6 22:29:55 2012] Using pre-built transcriptome index..
[Mon Feb 6 22:29:55 2012] Mapping left_kept_reads against transcriptome known with Bowtie
[Mon Feb 6 23:19:26 2012] Converting left_kept_reads.m2g to genomic coordinates (map2gtf)
[Mon Feb 6 23:54:00 2012] Resuming TopHat pipeline with unmapped reads
[Mon Feb 6 23:54:00 2012] Mapping left_kept_reads.m2g_um against hg19 with Bowtie
[Tue Feb 7 01:08:53 2012] Processing bowtie hits
[Tue Feb 7 01:25:39 2012] Mapping left_kept_reads.m2g_um_seg1 against hg19 with Bowtie (1/2)
[Tue Feb 7 02:20:36 2012] Mapping left_kept_reads.m2g_um_seg2 against hg19 with Bowtie (2/2)
[Tue Feb 7 03:06:23 2012] Searching for junctions via segment mapping
[Tue Feb 7 04:43:56 2012] Retrieving sequences for splices
[Tue Feb 7 05:08:27 2012] Indexing splices
[Tue Feb 7 05:19:54 2012] Mapping left_kept_reads.m2g_um_seg1 against segment_juncs with Bowtie (1/2)
[Tue Feb 7 05:49:43 2012] Mapping left_kept_reads.m2g_um_seg2 against segment_juncs with Bowtie (2/2)
[Tue Feb 7 06:17:11 2012] Joining segment hits
[Tue Feb 7 06:36:14 2012] Reporting output tracks