v3.4 Exomiser issue

[abrusell]

Hi Véronique,
Thanks for the great work.
I've installed v3.4, and I'm having issues with exomiser and phenogenius.
Specifying the GRCh38 build in the configfile (-genomeBuild: GRCh38) and in the command line, I get the following error in the log exomiser.tmp file:

2024-02-29 16:37:19.192  INFO 2155 --- [           main] com.zaxxer.hikari.HikariDataSource       : exomiser-genome-hg19-2309 - Starting...
2024-02-29 16:37:20.195 ERROR 2155 --- [           main] com.zaxxer.hikari.pool.HikariPool        : exomiser-genome-hg19-2309 - Exception during pool initialization.

org.h2.jdbc.JdbcSQLException: Database "/g100_work/OPBG2_prod2023/NGS_tools/ExomiserData/data/2309_hg19/2309_hg19_genome" not found [90013-197]

I've tried to change "exomiser.hg19.data-version=2309" to "exomiser.hg38.data-version=2309" in the configfile
and now it doesn't complain anymore but the column "Exomiser_gene_pheno_score" in the output file reports only "0" or "-1" values

Furthermore, I've correctly installed PhenoGenius in the $AnnotSV_3.4/share/python3/phenogenius directory, but I get this error:

...running PhenoGenius for 4524 gene names (March 01 2024 - 15:37)

child process exited abnormally

Any hints on how to solve these errors?

Thanks

[lgmgeo]

I am out of the office. I will be back on March 11 and will respond to your issue at that time.

[lgmgeo]

Ok, 2 problems here:

• Exomiser
• PhenoGenius
  I open another issue so as not to mix everything up.
  cf PhenoGenius issue #218

Let's go here with the Exomiser issue.

First of all, it should be noted that the Exomiser analyzes are independent of the genome build (cf #12).
The Exomiser module only uses the name of the genes (and not the position of the genes).
=> So don't worry about seeing "hg19" in the output.

The prolem is:

org.h2.jdbc.JdbcSQLException: Database "/g100_work/OPBG2_prod2023/NGS_tools/ExomiserData/data/2309_hg19/2309_hg19_genome" not found [90013-197]

Can you give me your exact AnnotSV command line?
Did you use the -annotationsDir option?
Did you run AnnotSV using Bioconda?

Because in a standard install, the exomiser database is located in:
$ANNOTSV/share/AnnotSV/Annotations_Exomiser/2309/2309_hg19/2309_hg19_genome.h2.db

As an example, the -annotationsDir option can be set to $ANNOTSV/share/AnnotSV.
WARNING: the directory architecture must remain the same as in AnnotSV:

ls $ANNOTSV/share/AnnotSV
Annotations_Exomiser  
Annotations_Human  
jar

ls $ANNOTSV/share/AnnotSV/Annotations_Exomiser/
2309

[abrusell]

So I've restored the $ANNOTSV/share/AnnotSV/Annotations_Exomiser/2309/2309_hg19/2309_hg19_genome.h2.db file and everything works!

Thank you!

[poddarharsh15]

Hi @lgmgeo,

Apologies for bringing this up here, but I believe it will be easier to clarify. I plan to use the 2402_hg38/ dataset for Exomiser instead of 2402_hg19/, as all of my VCF files are aligned to GRCh38. Would it be sufficient to simply change exomiser.hg19.data-version=2402 to exomiser.hg38.data-version=2402 in the applications.config file? Will this allow Exomiser to process the GRCh38 data correctly?

Thank you for your guidance.

[lgmgeo]

The Exomiser analyzes are independent of the genome build (cf #12).
The Exomiser module only uses the name of the genes (and not the position of the genes).
The 2402_hg19 file is OK for hg38 analysis, but it is to notice that the latest version of Exomiser officially supported by AnnotSV is 2309. I have never tested Exomiser version 2406.

[julesjacobsen]

@lgmgeo - have you tried updating your version of the exomiser-rest-prioritiser to exomiser-rest-prioritiser-14.1.0.jar? This doesn't need the spurious exomiser.hg19.data-version=2309 in the application.properties and it only requires the 2410_phenotype data to run.

[lgmgeo]

Hi @julesjacobsen, thanks for this information.
I haven't reviewed this version, I'll keep that in mind for a near future.