Cuffmerge error when merging transcript.gtf files

Question

[sarai]

Hi everyone!

I'm new on RNA-Seq analysis and I need your help. I'm trying to merge the transcript.gtf files I got with cufflinks but when I write the following command I get an error.

Can anyone suggest me how to solve this?

Thanks so much!

cuffmerge -g Mus_musculus.GRCm38.101.gtf.gz -s Mus_musculus.GRCm38.dna.primary_assembly.fa.gz assemblies.txt

[Sat Dec  5 20:19:09 2020] Beginning transcriptome assembly merge
-------------------------------------------

[Sat Dec  5 20:19:09 2020] Preparing output location ./merged_asm/
Traceback (most recent call last):
  File "/home/msmenfig/tools/cufflinks-2.2.1.Linux_x86_64/cuffmerge", line 580, in <module>
    sys.exit(main())
  File "/home/msmenfig/tools/cufflinks-2.2.1.Linux_x86_64/cuffmerge", line 546, in main
    chrom_info = get_gtf_chrom_info(gtf, chrom_info)
  File "/home/msmenfig/tools/cufflinks-2.2.1.Linux_x86_64/cuffmerge", line 476, in get_gtf_chrom_info
    left = int(cols[3])
ValueError: invalid literal for int() with base 10: '9\x11\x9ei;\xb6}\xf6\xce\x08\xbe\x90\x8d;\xf8\x95\x8bS\xb5\xa6\xb3q\xb7\xf6\xcc\xea\x1fO\xcf\x9fO_9\xa9\xaa2?\x18X\xe4{\xa5\x97\x1d\xa3rRR\x9a>\x93H\xce\x82\xe3\x1b;\x12\xcf\xee)\x99\xb7`C\x9f+\x95[7\x18\x94\x8f\xe1?0\xc3.f\xa6\x90\xa5W\xf8\x15\x8a\xcb]~E\x90\x0c\xcd\x07\x0fs\xbcJ\x9dl\xc8\xd1\x82\x8c\xd9JNd\xac\x0c\x05[\xb6k\xab\x1c"\xe3\xb2\x1803`\x07M\x86$2\xb6i\x866rW\xf0WF\x9d\x05\xa7\xc4\xd8\x8bnP\x8e\xe1\xc7\x1a|\x08%V\xf74\xe9'

[Ram]

Please use the formatting bar (especially the code option) to present your post better. You can use backticks for inline code (`text` becomes text), or select a chunk of text and use the highlighted button to format it as a code block. I've done it for you this time.

[Ram]

It looks like your input files contain weird characters. What is the output to:

file Mus_musculus.GRCm38.101.gtf.gz
file Mus_musculus.GRCm38.dna.primary_assembly.fa.gz

Also, ensure that cuffmerge can indeed work with gzipped input files.

[sarai]

What do you mean with "the output to" the files?

[Ram]

Those are commands, not files.

[sarai]

file Mus_musculus.GRCm38.101.gtf.gz 
Mus_musculus.GRCm38.101.gtf.gz: gzip compressed data, from Unix
file Mus_musculus.GRCm38.dna.primary_assembly.fa.gz
Mus_musculus.GRCm38.dna.primary_assembly.fa.gz: gzip compressed data

[Ram]

It looks like there might be a problem with the GTF file. Does cuffmerge accept a gzipped GTF file? Try running

cuffmerge -g <(zcat Mus_musculus.GRCm38.101.gtf.gz) -s Mus_musculus.GRCm38.dna.primary_assembly.fa.gz assemblies.txt

If that errors out too, try

cuffmerge -g <(zcat Mus_musculus.GRCm38.101.gtf.gz) -s <(zcat Mus_musculus.GRCm38.dna.primary_assembly.fa.gz) assemblies.txt

[sarai]

I got this with the second command line:

cuffmerge -g <(zcat Mus_musculus.GRCm38.101.gtf.gz) -s <(zcat Mus_musculus.GRCm38.dna.primary_assembly.fa.gz) assemblies.txt

[Sat Dec  5 22:52:30 2020] Beginning transcriptome assembly merge

-------------------------------------------

[Sat Dec  5 22:52:30 2020] Preparing output location ./merged_asm/
[Sat Dec  5 22:52:43 2020] Converting GTF files to SAM
[22:52:43] Loading reference annotation.
[22:52:46] Loading reference annotation.
[22:52:49] Loading reference annotation.
[22:52:51] Loading reference annotation.
[22:52:54] Loading reference annotation.
[22:52:57] Loading reference annotation.
[22:53:00] Loading reference annotation.
[22:53:04] Loading reference annotation.
[22:53:06] Loading reference annotation.
[Sat Dec  5 22:53:11 2020] Quantitating transcripts
Warning: Could not connect to update server to verify current version. Please check at the Cufflinks website (http://cufflinks.cbcb.umd.edu).
Command line:
cufflinks -o ./merged_asm/ -F 0.05 -g /dev/fd/63 -q --overhang-tolerance 200 --library-type=transfrags -A 0.0 --min-frags-per-transfrag 0 --no-5-extend -p 1 ./merged_asm/tmp/mergeSam_fileBPezAf 
[bam_header_read] EOF marker is absent. The input is probably truncated.
[bam_header_read] invalid BAM binary header (this is not a BAM file).
File ./merged_asm/tmp/mergeSam_fileBPezAf doesn't appear to be a valid BAM file, trying SAM...
[22:53:11] Loading reference annotation.
[22:53:11] Inspecting reads and determining fragment length distribution.
Processed 43499 loci.                       
> Map Properties:
>   Normalized Map Mass: 1151266.00
>   Raw Map Mass: 1151266.00
>   Fragment Length Distribution: Truncated Gaussian (default)
>                 Default Mean: 200
>              Default Std Dev: 80
[22:53:13] Assembling transcripts and estimating abundances.
Processed 43499 loci.                       
[Sat Dec  5 22:59:05 2020] Comparing against reference file /dev/fd/63
Warning: Could not connect to update server to verify current version. Please check at the Cufflinks website (http://cufflinks.cbcb.umd.edu).
Error: fasta file /dev/fd/62 not found!
    [FAILED]
Error: could not execute cuffcompare

And I get the directory merged_asm with these files:

genes.fpkm_tracking  isoforms.fpkm_tracking  logs  skipped.gtf  tmp  transcripts.gtf

[Ram]

I think it failed because it released the FASTA prematurely. Try re-running it, but this time gunzip the fasta file first.

gunzip Mus_musculus.GRCm38.dna.primary_assembly.fa.gz
cuffmerge -g <(zcat Mus_musculus.GRCm38.101.gtf.gz) -s Mus_musculus.GRCm38.dna.primary_assembly.fa assemblies.txt

[sarai]

I got the following. Should I assume this time it worked since I got the merged.gtf file?

cuffmerge -g <(zcat Mus_musculus.GRCm38.101.gtf.gz) -s Mus_musculus.GRCm38.dna.primary_assembly.fa assemblies.txt

[Sat Dec  5 23:36:21 2020] Beginning transcriptome assembly merge
-------------------------------------------

[Sat Dec  5 23:36:21 2020] Preparing output location ./merged_asm/
[Sat Dec  5 23:36:36 2020] Converting GTF files to SAM
[23:36:36] Loading reference annotation.
[23:36:39] Loading reference annotation.
[23:36:41] Loading reference annotation.
[23:36:44] Loading reference annotation.
[23:36:47] Loading reference annotation.
[23:36:50] Loading reference annotation.
[23:36:53] Loading reference annotation.
[23:36:57] Loading reference annotation.
[23:36:59] Loading reference annotation.
[Sat Dec  5 23:37:03 2020] Quantitating transcripts
Warning: Could not connect to update server to verify current version. Please check at the Cufflinks website (http://cufflinks.cbcb.umd.edu).
Command line:
cufflinks -o ./merged_asm/ -F 0.05 -g /dev/fd/63 -q --overhang-tolerance 200 --library-type=transfrags -A 0.0 --min-frags-per-transfrag 0 --no-5-extend -p 1 ./merged_asm/tmp/mergeSam_file43EPNZ 
[bam_header_read] EOF marker is absent. The input is probably truncated.
[bam_header_read] invalid BAM binary header (this is not a BAM file).
File ./merged_asm/tmp/mergeSam_file43EPNZ doesn't appear to be a valid BAM file, trying SAM...
[23:37:03] Loading reference annotation.
[23:37:03] Inspecting reads and determining fragment length distribution.
Processed 43499 loci.                       
> Map Properties:
>   Normalized Map Mass: 1151266.00
>   Raw Map Mass: 1151266.00
>   Fragment Length Distribution: Truncated Gaussian (default)
>                 Default Mean: 200
>              Default Std Dev: 80
[23:37:05] Assembling transcripts and estimating abundances.
Processed 43499 loci.                       
[Sat Dec  5 23:43:57 2020] Comparing against reference file /dev/fd/63
Warning: Could not connect to update server to verify current version. Please check at the Cufflinks website (http://cufflinks.cbcb.umd.edu).
No fasta index found for Mus_musculus.GRCm38.dna.primary_assembly.fa. Rebuilding, please wait..
Fasta index rebuilt.
[Sat Dec  5 23:44:47 2020] Comparing against reference file /dev/fd/63
Warning: Could not connect to update server to verify current version. Please check at the Cufflinks website (http://cufflinks.cbcb.umd.edu).
[msmenfig@node157 fastq_files]$ cd merged_asm
[msmenfig@node157 merged_asm]$ ls
logs  merged.gtf

[Ram]

Yes, I think so. More importantly, you did not get any errors. I think cuffmerge failed the last time because it cannot look for an index file when the file descriptor passed to it is a /dev/fd/ and not an actual filename.