NCI DTP Molecular Target Query Skill

SkillDatabases & data

Query the NCI-60 Molecular Target (Protein) database from the Developmental Therapeutics Program. Use when the user asks about protein expression of drug targets across the NCI-60 cancer cell line panel, or wants to look up a gene, cell line, or cancer panel in the NCI DTP molecular target dataset.

Instructions available. Your AI can read the instructions. Execution depends on the setup they require.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the NCI DTP Molecular Target Query Skill skill

What this skill tells your AI

The instructions your AI receives, as published by qsong-github/drugclaw in skills/dti/molecular_targets_data/SKILL.md and read by ahel’s review.

Search NCI-60 protein-level molecular target data by any entity. Auto-detects type:

Input PatternDetected AsMatch Logic
EGFR, TP53gene / proteinexact on GENE or substring on ENTITY_MEASURED, TITLE
MCF7, NCI-H460cell linesubstring on cellname
Breast, Leukemiacancer panelsubstring on pname
12345 (pure digits)MOLTIDexact on MOLTID (NCI pattern #)

Drug-Relevance Guide

Only protein expression data is downloaded — proteins are the direct molecular targets of drugs (kinase inhibitors → kinase expression, antibodies → receptor levels).

DatasetDownload?Reason
WEB_DATA_PROTEIN.ZIPYESDrug targets: protein expression across NCI-60
WEB_DATA_ALL_MT.ZIPOptionalSuperset incl. enzyme activity (also drug-relevant)
WEB_DATA_DNA.ZIPNoGenomic characterisation, not drug targets
WEB_DATA_SEQUENOM_METHYLATION.ZIPNoEpigenetic; indirect
WEB_DATA_*_MIR.ZIPNomicroRNA regulation; indirect
WEB_DATA_METABOLON*.ZIPNoMetabolomics; downstream, indirect
Microarray / SNP / CopyNum / KaryotypeNoCell-line genomic profiling; indirect

API

FunctionInputReturns
load_data(path)file or directory pathlist[dict]
search(data, entity)data + single entity stringlist[dict]
search_batch(data, entities)data + list of entity stringsdict[str, list[dict]]
summarize(hits, entity)hit list + labelcompact LLM-readable text
to_json(hits)hit listlist[dict] (JSON-serialisable)
query(data, entities, top_n)data + str or listtext block

Record Fields

Each record contains:

FieldDescription
MOLTIDNCI pattern number (molecular target ID)
GENEGene symbol
TITLEGene / protein full name
MOLTNBRNCI experiment ID
PANELNBR / CELLNBRInternal panel / cell identifiers
pnameCancer panel name (e.g., Breast, Leukemia)
cellnameCell line name (e.g., MCF7, A549/ATCC)
ENTITY_MEASUREDWhat was measured (e.g., protein name)
GeneIDNCBI Gene ID
UNITSMeasurement units
METHODAssay method
VALUENumeric measurement value
TEXTAdditional notes

Usage

See if __name__ == "__main__" block in 16_NCI_DTP_MolTarget.py for runnable examples covering: single gene query, cell line profile, batch multi-target search, cancer panel query, and JSON output.

Quick Examples

from importlib.machinery import SourceFileLoader
mt = SourceFileLoader("mt", "16_NCI_DTP_MolTarget.py").load_module()

data = mt.load_data()

# Single drug target
print(mt.query(data, "EGFR"))

# Multiple targets
print(mt.query(data, ["TP53", "BRAF", "HER2"]))

# Cell line molecular profile
print(mt.query(data, "MCF7"))

# JSON for downstream pipeline
import json
hits = mt.search(data, "EGFR")
print(json.dumps(mt.to_json(hits[:5]), indent=2))

Data Source & Download

Download Commands

# File already at:
# resources_metadata/dti/Molecular Target Data/WEB_DATA_PROTEIN.TXT

Signals

GitHub stars
116
Forks
3
Last commit
Aug 2026
Advanced
Item type
skill
Key
nci-dtp-moltarget
Source
github.com/qsong-github/drugclaw