Mosaic is now liveMCP tools, 760+ targets across oncology, neuro and cardio, preprint on Zenodo.See plans →
Mosaic
All theses

Target thesis · Metabolic enzyme (neomorphic gain-of-function)

IDH1

Isocitrate dehydrogenase 1

Generated 2026-07-12 · KG version 84 · every number traces to the evidence appendix

Coverage note

read this firstCoverage confidence: Partial

Partial coverage — some dimensions thin or reframed; caveats below.

DimensionHoldingTier
Literature134 curated papers (PubMed ~5,734 oncology)Partial
Patents101 patents, 139 organizationsHigh
Compounds72 distinct compounds (ChEMBL 337; edge-ratio 0.21)High
Genetic (DepMap)present, but dependency ≈ 0 — wrong lens (see Bear #1)Partial
Structuralpresent (Foldseek/AlphaFold + AlphaMissense)High

Coverage honesties, up front

  • Literature is Partial (98 papers) — directional, not deep.
  • DepMap “present” is misleading here: wild-type IDH1 is not a dependency (essential in 0.75% of models), because IDH1's actionable biology is the neomorphic mutant, not whole-gene essentiality. The dependency dimension is not used as a bull.

Assessed with confidence

  • Clinical validation
  • The neomorphic mechanism
  • The paralog-resistance theme

Flagged / withheld

  • Literature is Partial (98 papers — directional)
  • Genetic-dependency lens inapplicable (neomorphic mutant)

Target-level readiness: Thesis-ready.

Headline verdict

No strong angle· matches its reputation

A validated, approved target that the genetic-dependency lens completely misses. DepMap shows essentially no dependency (0.75% of models), yet IDH1 has approved drugs (max phase 4) — because the target is a neomorphic gain-of-function: the mutant enzyme (R132-region) produces the oncometabolite 2-HG, and the drugs hit the mutant, not wild-type essentiality. Read DepMap as the wrong instrument here. There is no strong non-consensus angle; the documented resistance theme is the IDH2 paralog.

The bull case

  1. 1.

    Clinically validated — drugs are approved. Max clinical phase 4; 72 compounds (phase distribution {phase-4: 3, phase-2: 3, phase-1: 2, preclinical: 64}); ChEMBL edge-ratio 0.21. The mechanism works in the clinic.

    confidence: highcompetitive_landscape

  2. 2.

    Sharp, hotspot-driven mutational landscape. AlphaMissense flags 61.4% of 4,541 variants pathogenic, hotspot at residue 137 — the catalytic-site region where the neomorphic mutations cluster.

    confidence: highgenetic_validation.variant_pathogenicity

  3. 3.

    Concentrated, committed competition. 101 patents / 139 organizations, led by Servier and Geron — established owners, not speculative.

    confidence: moderatecompetitive_landscape

The bear case

  1. 1.

    The genetic-dependency lens is inapplicable — read it correctly. DepMap mean gene effect -0.0954, essential in only 0.75% of 1,208 models. Wild-type IDH1 is not a fitness gene; the target is a neomorphic mutant (corroborated by the IDH2 resistance snippet — “Neomorphic isocitrate dehydrogenase (IDH) mutations lead to...”), so whole-gene knockout cannot evidence it. Efficacy is restricted to IDH1-mutant tumors.

    confidence: highgenetic_validation + resistance_bypass_map

  2. 2.

    Mutant-restricted indication. The opportunity exists only where the IDH1 mutation is present — a narrow, biomarker-gated population.

    confidence: highgenetic_validation

  3. 3.

    Literature is only Partial. 134 curated papers — assess directionally, not deeply.

    confidence: flaggedcoverage.dimensions.literature

  4. 4.

    Paralog resistance. The documented escape/related node is the IDH2 paralog (2 papers; also a neomorphic, drugged enzyme) — a known cross-axis consideration.

    confidence: moderateresistance_bypass_map

The non-consensus angleNo strong angle

No strong non-consensus signal in current data; IDH1's profile matches its reputation as a neomorphic, mutant-selective target. No DepMap co-essentiality fired.

  • A TCA-cycle neighbourhood, but PPI-proxy only. The functional-coupling whitespace tool returns ACO1, ACO2, IDH3G, IDH3B, CS, OGDH (PPI 0.99+) — the citric-acid-cycle enzymes around IDH1. None is in Mosaic's curated target set, so the graph has no patent/compound coverage for them; their competitive status is not assessed (not a whitespace claim). Coherent metabolism, but PPI-proxy, not co-essentiality, and not advanced.

    confidence: PPI-proxy — weakkg_native.synthetic_lethal_whitespace (PPI-proxy)

    Patent / compound status: not assessed (out of coverage). gene outside Mosaic curated coverage — competitive landscape is a validated lead to verify externally, not a whitespace claim.

  • IDH2 is the one evidenced resistance/related node and it is consensus (a drugged paralog), not whitespace.

    confidence: moderatekg_native.resistance_bypass_map (curated)

Honest empties

  • No co-essentiality fired.
  • emerging_signals shows no surge.
  • Modality near-uninformative (3 small-molecule typed of 72; 96% unclassified).

What would change this thesis

  • If a synthetic-lethal vulnerability of IDH1-mutant cells is evidenced by co-essentiality (a 2-HG-pathway collateral dependency), a real non-consensus angle emerges beyond direct mutant inhibition.
  • If resistance literature accumulates beyond the IDH2 paralog, the durability picture becomes assessable.
  • If literature coverage deepens past Partial, the assessment firms from directional to confident.