Humanbase Tools#

Configuration File: humanbase_tools.json Tool Type: Local Tools Count: 1

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

Available Tools#

humanbase_ppi_analysis (Type: HumanBaseTool)#

Retrieve a tissue-specific HumanBase gene interaction network and associated Gene Ontology biolog…

humanbase_ppi_analysis tool specification

Tool Information:

  • Name: humanbase_ppi_analysis

  • Type: HumanBaseTool

  • Description: Retrieve a tissue-specific HumanBase gene interaction network and associated Gene Ontology biological processes. Each edge includes datatype weights, the underlying HumanBase evidence-dataset records, their returned count, and an explicit completeness/error indicator. Top-level metadata distinguishes requested_scope_complete from globally exhaustive complete: HumanBase does not report whether more neighbors exist when the requested neighbor count is filled.

Parameters:

  • gene_list ([‘array’, ‘string’]) (optional) Genes to analyze, as a list of symbols or a comma-separated string. Official symbols and common synonyms both resolve (‘TP53’ or ‘p53’); the symbol actually used is reported back in resolved_genes.

  • tissue (string) (optional) Tissue type for tissue-specific interactions. Use simple names like ‘brain’, ‘heart’, ‘liver’, ‘kidney’, ‘lung’, ‘blood’. For v3 networks with more cell-type specificity, append ‘-v3’ (e.g., ‘brain-v3’, ‘liver-v3’).

  • max_node (integer) (optional) Legacy alias for neighbor_count: the number of interaction partners (additional neighboring genes) to retrieve, not the maximum total nodes. The network also contains the resolved query genes themselves, so it holds up to max_node + len(gene_list) proteins. Warning: the more nodes, the longer the retrieval.

  • interaction (string) (optional) Specific HumanBase datatype to use for network ranking/weights. If omitted, all seven upstream datatypes are requested, including protein-domain and sequence-similarity, which older wrapper versions silently omitted.

  • string_mode (boolean) (optional) If true, return a formatted network summary with evidence dataset counts; full dataset records remain available only in structured mode. If false, return a NetworkX graph whose edge attributes include evidence_datasets, evidence_dataset_count, evidence_datasets_complete, and evidence_error.

  • genes ([‘array’, ‘string’]) (optional) Alias for gene_list. A list of gene symbols or a comma-separated string.

  • top_n (integer) (optional) Legacy alias for neighbor_count. It requests this many interaction partners (additional neighboring genes), not this many total nodes.

  • neighbor_count (integer) (optional) Number of additional neighboring genes requested from HumanBase. Total returned nodes may be up to the unique resolved seed genes plus neighbor_count. The service provides no total or continuation token, so metadata.network_has_more is unknown when this count is filled. Defaults to 10 when no alias is supplied.

Example Usage:

query = {
    "name": "humanbase_ppi_analysis",
    "arguments": {
    }
}
result = tu.run(query)