ClinVar Variant ACMG Evidence Prediction Results

This CSV file contains predictions for three types of evidence (functional, population, and computational) for all ClinVar variant submissions.

The output CSV contains the following columns:
Identifier Columns
SCV: Submission accession number from ClinVar (format: SCV000000000)
VCV: Variation accession number from ClinVar (format: VCV000000000)
RCV: Record accession number from ClinVar (format: RCV000000000)
VariationID: Numerical identifier for the genetic variation

Genomic Coordinates
GRCh38_Chr: Chromosome number
GRCh38_Start: Start position on chromosome (GRCh38/hg38 assembly)
GRCh38_Stop: Stop position on chromosome (GRCh38/hg38 assembly)
GRCh38_ReferenceAllele: Reference allele sequence
GRCh38_AlternateAllele: Alternate allele sequence

Protein-Level Information
aapos: Amino acid position in the protein
aaref: Reference amino acid (single letter code)
aaalt: Alternate amino acid (single letter code)
gene: Gene symbol (e.g., BRCA1, TP53)

Original Classification
SubmissionClassification: Original classification provided by the submitter mainly includes pathogenic, likely pathogenic, uncertain significance, benign, likely benign, but also includes other values

Input Text
Comment: The textual comment/description provided with the variant submission that was used as input to the model

Model Predictions
has_evidence: does the text summary contain a specific type of evidence or not (TRUE or FALSE)
evidence_confidence: model prediction confidence (0.0 to 1.0)
predicted_evidence: if the text summary contains a specific type of evidence, model prediction of whether it is pathogenic or benign (P or B for population or computational evidence, PS3 or BS3 for functional evidence)
B_Score: model prediction confidence for being a benign type of evidence (0.0 to 1.0)
P_Score: model prediction confidence for being a pathogenic type of evidence (0.0 to 1.0)

Notes on Probability Scores
Probability scores (B_Score, P_Score) sum to 1.0 for each row. Higher probability indicates greater model confidence for that classification. The predicted_evidence corresponds to the classification with the highest probability score.

Model Type: Fine-tuned transformer model for sequence classification
Input: Textual comments from variant submissions
Training Data: ClinVar variant submissions with known classifications and mentions of ACMG evidence guideline