I am using starchip's fusion mode to detect integration sites between the host (human) and viral genomes that I know the host is infected with. To allow starchip to recognize the fusion pair between the host and viral genome, I modified both the reference fasta that STAR was run with and the HG38 genome GTF file to add in entires for the virus as if part of the human reference (a la "a 25th chromosome"). From this point, I generated the BED files using the setup.sh script and have modified the fus_parameters only to point to the correct reference fasta and GTF files.
Running the fusion script, I detect several fusions; but upon entering the filter step, I get the following error:
***** ERROR: too many digits/characters for integer conversion in string . Exiting...
This ends the process there and the generated output_summary files are empty.
Error was due to incorrect construction of GTF file; an additional field was included before the chromosomal positions that screwed up the BED file creation.
I am using starchip's fusion mode to detect integration sites between the host (human) and viral genomes that I know the host is infected with. To allow starchip to recognize the fusion pair between the host and viral genome, I modified both the reference fasta that STAR was run with and the HG38 genome GTF file to add in entires for the virus as if part of the human reference (a la "a 25th chromosome"). From this point, I generated the BED files using the setup.sh script and have modified the fus_parameters only to point to the correct reference fasta and GTF files.
Running the fusion script, I detect several fusions; but upon entering the filter step, I get the following error:
This ends the process there and the generated output_summary files are empty.
Any idea why this might be happening?