search

suhrig / arriba

Fast and accurate gene fusion detection from RNA-Seq data
Other
226 stars 50 forks source link

EOF marker is absent. The input may be truncated #192

Closed picardda closed 1 year ago

picardda commented 1 year ago

Hello and thank you for generating this tool. I am new at the RNA sequencing alignments and fusion detection, but I am hoping to get your program to work.

I have installed arriba using the command on the website: wget https://github.com/suhrig/arriba/releases/download/v2.4.0/arriba_v2.4.0.tar.gz tar -xzf arriba_v2.4.0.tar.gz cd arriba_v2.4.0 && make

and then downloaded the reference: ./download_references.sh hs37d5viral+GENCODE19

Everything seems to be fine. However, when I run arriba as stated in the quickstart: ./run_arriba.sh STAR_index_hs37d5viral_GENCODE19/ GENCODE19.gtf hs37d5viral.fa database/blacklist_hg19_hs37d5_GRCh37_v2.4.0.tsv.gz database/known_fusions_hg19_hs37d5_GRCh37_v2.4.0.tsv.gz database/protein_domains_hg19_hs37d5_GRCh37_v2.4.0.gff3 8 test/read1.fastq.gz test/read2.fastq.gz

I get the following error: EXITING because of fatal PARAMETERS error: --chimMultimapNmax > 0 (new chimeric detection) presently only works with --chimOutType Junctions SOLUTION: re-run with --chimOutType Junctions

Mar 24 09:47:26 ...... FATAL ERROR, exiting [W::bgzf_read_block] EOF marker is absent. The input may be truncated (total=ERROR: no normal reads found

I tried to change the run_arriba.sh file to "--chimOutType Junctions", but I received an error that made me change it back.

Is there anything you could suggest I try? Not sure if it is important, but I am running STAR 2.7.3a. Thank you for your help, Daniel

suhrig commented 1 year ago

Your STAR version is too old. The minimum requirement is >= 2.7.6a. However, I recommend >=2.7.10a, because there are some good improvements for fusion detection coming with this version.