Spliceai Tools

Configuration File: spliceai_tools.json Tool Type: Local Tools Count: 3

This page contains all tools defined in the spliceai_tools.json configuration file.

Available Tools

SpliceAI_get_max_delta (Type: SpliceAITool)

Get simplified maximum SpliceAI delta score and pathogenicity interpretation for a variant. Retur…

SpliceAI_get_max_delta tool specification

Tool Information:

  • Name: SpliceAI_get_max_delta

  • Type: SpliceAITool

  • Description: Get simplified maximum SpliceAI delta score and pathogenicity interpretation for a variant. Returns single max score across all splice effects (AG/AL/DG/DL) with interpretation: benign (<0.2), low (0.2-0.5), moderate (0.5-0.8), high (≥0.8). Best for quick variant triage.

Parameters:

  • variant (string) (required) Variant in chr-pos-ref-alt format (e.g., chr8-140300616-T-G)

  • genome (string) (optional) Genome build: 37 or 38 (default: 38)

Example Usage:

query = {
    "name": "SpliceAI_get_max_delta",
    "arguments": {
        "variant": "example_value"
    }
}
result = tu.run(query)

SpliceAI_predict_pangolin (Type: SpliceAITool)

Predict splice-altering effects using Pangolin model (alternative to SpliceAI). Pangolin uses a d…

SpliceAI_predict_pangolin tool specification

Tool Information:

  • Name: SpliceAI_predict_pangolin

  • Type: SpliceAITool

  • Description: Predict splice-altering effects using Pangolin model (alternative to SpliceAI). Pangolin uses a different architecture and may provide complementary predictions. Returns splice effect scores per transcript. Useful for validation or when SpliceAI gives borderline results.

Parameters:

  • variant (string) (required) Variant in chr-pos-ref-alt format (e.g., chr8-140300616-T-G)

  • genome (string) (optional) Genome build: 37 or 38 (default: 38)

  • distance (integer) (optional) Distance parameter for model (default: 1000)

  • mask (boolean) (optional) Use masked scores

Example Usage:

query = {
    "name": "SpliceAI_predict_pangolin",
    "arguments": {
        "variant": "example_value"
    }
}
result = tu.run(query)

SpliceAI_predict_splice (Type: SpliceAITool)

Predict splice-altering effects using SpliceAI deep learning model. Returns delta scores for acce…

SpliceAI_predict_splice tool specification

Tool Information:

  • Name: SpliceAI_predict_splice

  • Type: SpliceAITool

  • Description: Predict splice-altering effects using SpliceAI deep learning model. Returns delta scores for acceptor gain (DS_AG), acceptor loss (DS_AL), donor gain (DS_DG), donor loss (DS_DL). Scores ≥0.2 suggest splice impact, ≥0.5 moderate, ≥0.8 high pathogenicity. Critical for intronic/exonic variant interpretation.

Parameters:

  • variant (string) (required) Variant in chr-pos-ref-alt format (e.g., chr8-140300616-T-G) or colon-separated

  • genome (string) (optional) Genome build: 37 (GRCh37/hg19) or 38 (GRCh38/hg38). Default: 38

  • distance (integer) (optional) Distance parameter for model (default: 50). Larger values capture more distal effects.

  • mask (boolean) (optional) Use masked scores (recommended for variant interpretation). Raw scores better for alternative splicing analysis.

Example Usage:

query = {
    "name": "SpliceAI_predict_splice",
    "arguments": {
        "variant": "example_value"
    }
}
result = tu.run(query)