How do I add custom population frequency data (e.g., ABraOM) in the .yml template?

[DanielBarachoFontes]

I'm a layman in bioinformatics, so I apologize in advance if this is an ignorant question!

I'm trying to use Exomiser with Brazilian clinical cases. Here in Brazil, when working with population-level variant frequencies, we commonly use ABraOM (Arquivo Brasileiro Online de Mutações, Online Archive of Brazilian Mutations).
Is it possible to integrate ABraOM data into Exomiser? I believe the relevant section would be the "frequencySources" field in the .yml template:

frequencySources: [
    UK10K,

    GNOMAD_E_AFR,
    GNOMAD_E_AMR,
  #  GNOMAD_E_ASJ,
    GNOMAD_E_EAS,
  #  GNOMAD_E_FIN,
    GNOMAD_E_NFE,
  #  GNOMAD_E_OTH,
    GNOMAD_E_SAS,

    GNOMAD_G_AFR,
    GNOMAD_G_AMR,
  #  GNOMAD_G_ASJ,
    GNOMAD_G_EAS,
  #  GNOMAD_G_FIN,
    GNOMAD_G_NFE,
  #  GNOMAD_G_OTH,
    GNOMAD_G_SAS
]

I understand I would need to link Exomiser to the ABraOM dataset somehow, but I'm not sure how to go about it. Any guidance would be greatly appreciated!

[julesjacobsen]

Hi @DanielBarachoFontes Yes, this is possible. You can add a LOCAL frequency data source. e.g.

frequencySources: [
    LOCAL,
    UK10K,
# and the rest...
]

Then in your application.properties file edit and uncomment this line, ensuring you have provided the correct genome assembly (i.e. make sure you use hg19 for GRCh37 or hg38 for GRCh38):

# local frequencies are required to be normalised in the same manner as the input VCF and frequency values must be percentages.
#exomiser.hg19.local-frequency-path=${exomiser.data-directory}/local/local_frequency_test_hg19.tsv.gz

You will probably need to post-process the file to fit this file format:

https://github.com/exomiser/Exomiser/blob/master/exomiser-core/src/test/resources/local_frequency_test.tsv

and then bgzip and tabix index the file:

 bgzip file > file.gz
 tabix -s1 -b2 -e2 -c# file.gz

You can test if it is working by just running the LOCAL frequency in a analysis and checking the data matches what you expect from the original data source.

[DanielBarachoFontes]

I am very grateful :)