MapSplice Error: filter alignment failed

Question

[liguanghaoli]

Hi, When I use MapSplice to do mapping in RNA seq data analysis. The error occurred and there was no bam file generated. The error information is listed as follow:

[Thu Jan 11 15:42:04 2018] Inspecting Bowtie index files
[Thu Jan 11 15:42:04 2018] Checking reference sequence length
[Thu Jan 11 15:42:29 2018] Checking consistency of Bowtie index and reference sequence
[Thu Jan 11 15:42:29 2018] Checking read format
[Thu Jan 11 15:44:54 2018] Running MapSplice multi-thread
[Thu Jan 11 17:42:09 2018] Generating junctions from sam file
[Thu Jan 11 17:43:07 2018] Filtering junction by min mis and min lpq
[Thu Jan 11 17:43:07 2018] Filtering junction by ROC argu noncanonical
[Thu Jan 11 17:43:08 2018] Generating synthetic junction sequences
[Thu Jan 11 17:43:27 2018] Checking Bowtie index files
[Thu Jan 11 17:43:27 2018] Building Bowtie index for junction synthetic sequence
[Thu Jan 11 17:43:47 2018] Running MapSplice multi-thread
[Thu Jan 11 20:25:13 2018] Running alignment handler
[MapSplice Running Failed]
Error: filter alignment failed

Is there anyone have encounted this error before? I have read the user guide of MapSplice and couldn't find the solution. I think there may be some problems in filtering unmapped reads but don't know how to solve it. Could you please help me? Thank you very much. Guanghao Li

[simo.aufiero]

Hi Guanghao Li, I had the same problem. I am running mapsplice using the genome reference mm10. I did the analysis in the past by using mm9 and mapsplice pipeline finished without any problem with that genome reference. What genome reference are you using?

Simona

[liguanghaoli]

Hi, Simona, I use hg19 as reference. And now I run MapSplice in another computer, the program work well. So I'm also comfused about the error.

[staypositive]

maybe you should send the data to mapsplice at netlab.uky.edu, they can help you

[spin]

Hello, same error message here but it is inconsistent. Some of my samples from the same batch run perfectly fine. I use GRCh38 as reference. Have you been able to figure out the problem?

Thanks J

Answer 1

[simo.aufiero]

Hi Spin, this is the command line that I used to perform the analysis:

python mapsplice.py -p 16 --bam -o mapsplice_out001 -c Chromosomes -x BowtieIndex/genome -1 paired-001.R1.fastq -2 paired-001.R2.fastq --gene-gtf gencode.vM16.annotation.gtf --fusion-non-canonical --filtering 1 --min-map-len  25 --min-fusion-distance 200

Below I reported the log file of the analysis. As you can see in my case the error generated when mapsplice tried to build bowtie index for fusion junction synthetic sequence. I sent an email to netlab ([e-mail]) asking how I could solve the problem. They suggested to reduce the false positives:

1. by increasing the --min-map-len
2. if 1 does not solve the problem, by turning off --fusion-non-canonical

By increasing the --min-map-len to 50 and by turning off --fusion-non-canonical the analysis finished without any problem. I hope this can help you.

Simona

-----------------------------------------------
[Tue Jan 30 08:51:54 2018] Beginning Mapsplice run (MapSplice v2.2.0)
[Tue Jan 30 08:51:54 2018] Bin directory: /scratch/1055713.batch1.lisa.surfsara.nl/MS22_mm10/bin/ 
[Tue Jan 30 08:51:54 2018] Preparing output location mapsplice_out002/
[Tue Jan 30 08:51:54 2018] Checking files or directory: paired-002.R1.fastq
[Tue Jan 30 08:51:54 2018] Checking files or directory: paired-002.R2.fastq
[Tue Jan 30 08:51:54 2018] Checking files or directory: Chromosomes/
[Tue Jan 30 08:51:54 2018] Checking Bowtie index files
[Tue Jan 30 08:51:54 2018] Inspecting Bowtie index files
[Tue Jan 30 08:51:54 2018] Checking reference sequence length
[Tue Jan 30 08:52:15 2018] Checking consistency of Bowtie index and reference sequence
[Tue Jan 30 08:52:15 2018] Checking read format
-----[Read Format: FASTQ]
-----[Read Type: Pair End]
-----[Total # Reads: 134960420]
-----[Max Read Length: 151]
-----[Min Read Length: 50]
-----[Max Quality Score: 74]
-----[Min Quality Score: 35]
-----[Quality Score Scale: Phred+33]
[Tue Jan 30 09:08:00 2018] Running MapSplice multi-thread
[Tue Jan 30 10:44:48 2018] Generating junctions from sam file
[Tue Jan 30 10:46:28 2018] Filtering junction by min mis and min lpq
[Tue Jan 30 10:46:31 2018] Filtering junction by ROC argu noncanonical
[Tue Jan 30 10:46:33 2018] Generating synthetic junction sequences
[Tue Jan 30 10:46:52 2018] Checking Bowtie index files
[Tue Jan 30 10:46:52 2018] Building Bowtie index for junction synthetic sequence
[Tue Jan 30 10:47:21 2018] Running MapSplice multi-thread
[Tue Jan 30 14:33:45 2018] Running alignment handler
[Tue Jan 30 15:03:37 2018] Parsing cluster regions
[Tue Jan 30 15:03:51 2018] Clustering regions
[Tue Jan 30 15:03:57 2018] Running MapSplice multi-thread fusion
[Tue Jan 30 16:32:55 2018] Generating fusion junctions from sam file and filter by anchor
[Tue Jan 30 17:54:46 2018] Filtering original fusion junction
[Tue Jan 30 17:57:56 2018] Synthetic fusion junctions sequence
[Wed Jan 31 14:26:09 2018] Checking Bowtie index files
[Wed Jan 31 14:26:09 2018] Building Bowtie index for fusion junction synthetic sequence
Error: Reference sequence has more than 2^32-1 characters!  Please divide the
reference into batches or chunks of about 3.6 billion characters or less each
and index each independently.
Command: /scratch/1055713.batch1.lisa.surfsara.nl/MS22_mm10/bin/bowtie-build mapsplice_out002/tmp/fusion/fusion_synthetic_sequence.txt mapsplice_out002/tmp/fusion/syn_fusion_idx_prefix 
[MapSplice Running Failed]
Error: Splice sequence indexing failed
cp: cannot stat 'circular_RNAs.txt': No such file or directory
cp: cannot stat 'stats.txt': No such file or directory