Hello there,
I would like to test hera (https://github.com/bioturing/hera) for RNA-Seq analysis including virus-host fusion transcripts. Therefore I need to generate a new index. When I try to do this, I end up without error message but also with empty index files:
$ hera_build --fasta /path/to/hg38_w_vector.fa --gtf /path/to/hg38_ucsc_w_vector.gtf --outdir /path/to/hera_index/ --full_index 1
Scan reference sequence ...
Scan gtf file...
Extract transcript sequence
Hera is a program developed by BioTuring for RNA-Seq analysis.
Please contact info@bioturing.com if you need further support
Index transcripts in: /path/to/hera_index/transcripts.fasta
Number of sequences : 0
When I first tried the index building, there was an error because of missing os and os.path, respectively. After adding "import os" to the hera_build script, it ran through without error but as I've shown without correct index. I've tried with and without --grch38 1 with no difference but size of reference.inf (2 vs 25 kb).
Hera was installed via miniconda/bioconda without problems. Hera itself seems to run fine.
The server uses CentOS Linux release 7.2.1511.
Would be great, if anyone could help me with that.
Looks like I had the wrong gtf file for my genome fasta file. I tested with the GRCh38.fa and GRCh38.82.gtf using the link from https://github.com/bioturing/hera and now it seems to work.