Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →AI-powered analysis of chromosomal instability (CIN) signatures for cancer prognosis, immunotherapy response prediction, and therapeutic vulnerability identification.
.claude/skills/chromosomal-instability-agent/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 1% | 0% |
| case-14 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-02 | ✓→✓ | ＝ Same ✓ | 28% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | -32% | 0% |
<!--
#
#
-->
The Chromosomal Instability Agent analyzes CIN signatures to predict cancer prognosis, immunotherapy response, and therapeutic vulnerabilities. It integrates copy number alterations, aneuploidy scores, and CIN-related gene expression for comprehensive genomic instability assessment.
| Metric | Calculation | Interpretation | |--------|-------------|----------------| | Aneuploidy score | Arm-level alterations | Chromosome-level CIN | | SCNA burden | Total CNV alterations | Overall instability | | Weighted GII | Fraction altered genome | Focal vs broad changes | | CIN70 | 70-gene signature | Transcriptional CIN | | WGII | Weighted genome instability | Comprehensive score |
Core genes reflecting CIN phenotype:
User: "Analyze chromosomal instability in this breast cancer sample and identify treatment vulnerabilities."
Agent Action:
bashpython3 Skills/Oncology/Chromosomal_Instability_Agent/cin_analyzer.py \ --cnv_segments tumor_cnv.tsv \ --expression rnaseq_tpm.tsv \ --mutations somatic.maf \ --tumor_type breast_cancer \ --signatures cin70,cin25 \ --output cin_report/
High CIN Associates With:
Mechanisms:
| Target | Agents | CIN Context | |--------|--------|-------------| | PARP | Olaparib, etc. | High CIN + HRD | | ATR | Berzosertib | Replication stress | | WEE1 | Adavosertib | G2/M dependency | | CHK1 | Prexasertib | Cell cycle checkpoint | | KIF11 | Ispinesib | Mitotic dependency | | Aurora kinases | Alisertib | Mitotic errors |
| CIN Level | Prognosis | ICI Response | Alternative Therapy | |-----------|-----------|--------------|---------------------| | Low | Better | Better | Standard care | | Intermediate | Variable | Variable | Combination therapy | | High | Poor | Poor | CIN-targeted agents | | Extreme | Very poor | Immune desert | Chemotherapy |
CIN Score Prediction:
Prognosis Modeling:
Therapeutic Matching:
| Cancer Type | Typical CIN Level | Driver Events | |-------------|-------------------|---------------| | Ovarian HGSOC | Very high | TP53, BRCA | | Triple-neg breast | High | TP53, PI3K | | Colorectal MSS | Moderate-high | APC, TP53 | | Colorectal MSI | Low | MMR deficiency | | Thyroid (PTC) | Low | BRAF, RAS | | Melanoma | Moderate | BRAF, NRAS |
AI Group - Biomedical AI Platform
<!-- AUTHOR_SIGNATURE: 9a7f3c2e-MD-BABU-MIA-2026-MSSM-SECURE -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 23,298 | 21,650 | -7% | 1 | 1 | 0% | 4,451 | 6,256 | +41% | 0 | 0 | — |
case-02 | pass→pass | 12,497 | 8,109 | -35% | 1 | 1 | 0% | 2,333 | 2,978 | +28% | 0 | 0 | — |
case-03 | fail→pass | 19,671 | 7,941 | -60% | 1 | 1 | 0% | 2,563 | 3,036 | +18% | 0 | 0 | — |
case-04 | pass→pass | 17,538 | 3,660 | -79% | 1 | 1 | 0% | 3,191 | 2,171 | -32% | 0 | 0 | — |
case-05 | pass→pass | 8,902 | 7,107 | -20% | 1 | 1 | 0% | 1,766 | 2,745 | +55% | 0 | 0 | — |
case-06 | pass→pass | 12,711 | 11,540 | -9% | 1 | 1 | 0% | 2,134 | 3,349 | +57% | 0 | 0 | — |
case-07 | pass→pass | 14,050 | 14,826 | +6% | 1 | 1 | 0% | 2,529 | 4,147 | +64% | 0 | 0 | — |
case-08 | pass→pass | 11,087 | 10,746 | -3% | 1 | 1 | 0% | 2,051 | 3,313 | +62% | 0 | 0 | — |
case-09 | pass→pass | 12,383 | 10,232 | -17% | 1 | 1 | 0% | 2,140 | 3,201 | +50% | 0 | 0 | — |
case-10 | pass→pass | 9,852 | 5,326 | -46% | 1 | 1 | 0% | 1,644 | 2,398 | +46% | 0 | 0 | — |
case-11 | fail→pass | 13,971 | 5,338 | -62% | 1 | 1 | 0% | 2,327 | 2,360 | +1% | 0 | 0 | — |
case-12 | pass→pass | 8,848 | 7,680 | -13% | 1 | 1 | 0% | 1,472 | 2,806 | +91% | 0 | 0 | — |
case-13 | pass→pass | 7,332 | 6,023 | -18% | 1 | 1 | 0% | 1,379 | 2,470 | +79% | 0 | 0 | — |
case-14 | fail→pass | 19,597 | 11,320 | -42% | 1 | 1 | 0% | 3,395 | 3,466 | +2% | 0 | 0 | — |
case-15 | pass→pass | 13,072 | 8,992 | -31% | 1 | 1 | 0% | 2,377 | 2,934 | +23% | 0 | 0 | — |
case-16 | pass→pass | 12,798 | 12,400 | -3% | 1 | 1 | 0% | 2,299 | 3,472 | +51% | 0 | 0 | — |
case-17 | pass→pass | 11,858 | 7,851 | -34% | 1 | 1 | 0% | 2,073 | 2,790 | +35% | 0 | 0 | — |
case-18 | pass→pass | 11,921 | 8,643 | -27% | 1 | 1 | 0% | 1,996 | 2,913 | +46% | 0 | 0 | — |
case-19 | pass→pass | 12,386 | 10,938 | -12% | 1 | 1 | 0% | 2,326 | 3,310 | +42% | 0 | 0 | — |
case-20 | pass→pass | 13,377 | 18,602 | +39% | 1 | 1 | 0% | 2,517 | 5,356 | +113% | 0 | 0 | — |
case-21 | pass→pass | 9,291 | 23,036 | +148% | 1 | 1 | 0% | 1,625 | 4,360 | +168% | 0 | 0 | — |
case-22 | pass→pass | 11,414 | 8,671 | -24% | 1 | 1 | 0% | 2,153 | 3,180 | +48% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +14 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
| Model | Method | Date | Lift |
|---|---|---|---|
| gemini-3.6-flash | verified | 7/24/2026 | +18% |
Other measured skills in the registry, with their headline benchmark lift.