Error: GA bit thresholds unavailable on model RhiPru-1.174

[Bijie Ren]

Dear officer,

I'm trying to build a custom genome but when I tried to run the code “ perl dfamscan.pl -fastafile ref_annot.cdsplus.fa -hmmfile dfam_dbb/Dfam_db/Dfam.hmm -dfam_outfile __dfam_ref_annot.cdsplus.fa/dfam.out --masking_thresh --cpu 4”

there was an error: GA bit thresholds unavailable on model RhiPru-1.174

I'm wondering if you know what's the meaning of this?  Thank you so much for any info!

Best regards

Bijie Ren

[Brian Haas]

hi,

We tend to use an organism-specific dfam library here, such as human or mouse, instead of the largest Dfam.hmm, which tends to have problems.  Try finding the organism-specific version that might be the best fit for your target.

--
You received this message because you are subscribed to the Google Groups "STAR-Fusion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to star-fusion...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/star-fusion/CAM5Nko_fbJBuyfLac4Ym%3DUDOxmBJTmpgzjygzzYyGmMOZuZQKg%40mail.gmail.com.

--

--
Brian J. Haas
The Broad Institute
http://broadinstitute.org/~bhaas

 

[Bijie Ren]

Hi Brian,

Thank you so much!  I tried different sizes of Dfam (full or curated only) but failed to run the code again.   I also generated a blast-paired of my reference genome and would like to use --blast_pairs to build my custom genome index.  It will be like: 

prep_genome_lib_new.pl
--genome_fa genome.fa
--gtf genes_known_chrom.gtf
--blast_pairs blast_pairs.outfmt6.gz

However, it seems that the most updated version of prep_genome_lib.pl doesn't have the input parameter setting for --blast_pairs. I'm wondering if I can get the right version of prep_genome_lib.pl to run --blast_pairs instead of - dfam_db?

Best regards

Bijie Ren

[Brian Haas]

Hi Bijie,

The prep_genome_lib.pl script should be generating its own blast pairs file as part of the build process. It should require the dfam database to be provided - but don't use the Dfam.hmm - instead, use the organism-specific version that's most relevant to your target.  ie. there's a separate human, mouse, etc., for dfam that you can use.

best,

~brian

To view this discussion on the web visit https://groups.google.com/d/msgid/star-fusion/4c1916d4-0108-4b22-8080-e055b88069ben%40googlegroups.com.