Hi all,

I've got a set of interesting variants in a VCF file and I wanted to use alphagenome ( https://deepmind.google/blog/alphagenome-ai-for-better-understanding-the-genome/ ) to get a score. After calling the API for each variant, I get a large number of metrics for each variant. Looks like:

variant_id      scored_interval gene_id gene_name       gene_type       gene_strand     junction_Start  junction_End    output_type     variant_scorer  track_name      track_strand    Assay titleontology_curie   biosample_name  biosample_type  biosample_life_stage    data_source     endedness       genetically_modified    transcription_factor    histone_mark    gtex_tissue     raw_score  quantile_score
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000084 ATAC-seq      .       ATAC-seq        CL:0000084      T-cell  primary_cell    adult   encode  paired  False                           -0.0057604313   -0.28066123
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000100 ATAC-seq      .       ATAC-seq        CL:0000100      motor neuron    in_vitro_differentiated_cells   adult   encode  paired  False                           -0.007056713    -0.22671121
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000236 ATAC-seq      .       ATAC-seq        CL:0000236      B cell  primary_cell    adult   encode  paired  False                           -0.00051677227  -0.092026785
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000623 ATAC-seq      .       ATAC-seq        CL:0000623      natural killer cell     primary_cell    adult   encode  paired  False                           -0.004648924    -0.2699999
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000624 ATAC-seq      .       ATAC-seq        CL:0000624      CD4-positive, alpha-beta T cell primary_cell    adult   encode  paired  False                           -0.001062274    -0.11485085
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000625 ATAC-seq      .       ATAC-seq        CL:0000625      CD8-positive, alpha-beta T cell primary_cell    adult   encode  paired  False                           -0.00072705746  -0.10345244
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000787 ATAC-seq      .       ATAC-seq        CL:0000787      memory B cell   primary_cell    adult   encode  paired  False                           -0.0006222725   -0.080576845
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000788 ATAC-seq      .       ATAC-seq        CL:0000788      naive B cell    primary_cell    adult   encode  paired  False                           0.0007699728    -0.023067882
chr123:12345715190:C>CT chr123:12345190902-134239478:.                                                  ATAC    CenterMaskScorer(requested_output=ATAC, width=501, aggregation_type=DIFF_LOG2_SUM) CL:0000792 ATAC-seq      .       ATAC-seq        CL:0000792      CD4-positive, CD25-positive, alpha-beta regulatory T cell       primary_cell    adult   encode  paired  False                      -0.0018496513    -0.16011412
(...)

I'm a bit lost here (and the documentation is not clear to me) as there are 4620 distinct 'track_name'.

(...)
usage_UBERON:0011907 total RNA-seq
usage_UBERON:0012249 gtex Cervix_Ectocervix polyA plus RNA-seq
usage_UBERON:0013756 gtex Whole_Blood polyA plus RNA-seq
usage_UBERON:0015143 total RNA-seq
usage_UBERON:0018115 polyA plus RNA-seq
usage_UBERON:0018116 polyA plus RNA-seq
usage_UBERON:0018117 polyA plus RNA-seq
usage_UBERON:0018118 polyA plus RNA-seq
usage_UBERON:0036149 gtex Skin_Not_Sun_Exposed_Suprapubic polyA plus RNA-seq
usage_UBERON:0036149 total RNA-seq

how can I find the most 'interesting variant' whatever is the variant_scorer/track_name ? should I aggregate the data in some way ? which metrics should I use raw_score ? quantile_score ? can I use those metrics among all the type of analysis ?

thanks, P.