Q (sonnet-5.5): "Are mutations in CDKN2A, CDK4, and RB1 mutually exclusive in glioblastoma patients?" ▶ read_guide { "uri": "cbioportal://statistical-tests-guide" } ◀ result Statistical Test Selection Guide ================================ Purpose ------- This guide ensures the correct statistical test is selected before performing any group comparison, matching cBioPortal's own Group Comparison defaults. HARD RULES — NEVER FABRICATE A STATISTIC ---------------------------------------- ClickHouse cannot run statistical tests. The agent therefore must NEVER produce a derived statistic that is not a literal column value from a SQL result. Specifically: 1. **Never invent a p-value.** Not "p < 0.001", not "p ≈ 0.05", not any p-value. If the user asks "what is the p-value?", the answer is *"I can't compute that — here is the 2x2 contingency table (or group statistics). Run it in cBioPortal's Group Comparison tab, in R with `fisher.test(...)` / `wilcox.test(...)`, or in Python with `scipy.stats.fisher_exact(...)` / `mannwhitneyu(...)`."* 2. **Never claim mutual exclusivity (or co-occurrence) from a contingency table alone.** A 2x2 table is not a test. The shape "altered/not altered × group A/group B" needs Fisher's exact + a defined direction (odds ratio < 1 with significant p). Without that test, the agent presents the table and stops. Descriptive phrasing is also a claim: "largely/mostly mutually exclusive", "rarely co-occur", "tend to co-occur" are forbidden without the test — point to cBioPortal's Mutual Exclusivity tab instead. 3. **Never report a "median" that came from `AVG(...)` or any non-median aggregate.** "Median" and "mean" are different statistics; for skewed clinical distributions (especially survival) they differ substantially. Use ClickHouse's `quantile(0.5)(...)` for actual median, and label arithmetic averages as "mean", never "median". 4. **Never report a hazard ratio, odds ratio, risk ratio, or relative risk** that wasn't computed by an external tool. These require regression / model fitting that ClickHouse does not do. 5. **Never report median overall survival from `AVG(OS_MONTHS)` or even `quantile(0.5)(OS_MONTHS)`.** Median OS requires Kaplan-Meier estimation, which handles censoring (`OS_STATUS = 0:LIVING` means the event hasn't happened yet). Naive medians/means over `OS_MONTHS` ignore censoring and are systematically wrong. The correct handoff: return the raw `(OS_MONTHS, OS_STATUS)` pairs (or descriptive counts: N events, N censored, follow-up range) and tell the user to run KM in R (`survival::survfit`) or Python (`lifelines.KaplanMeierFitter`), or use cBioPortal's Survival comparison. If asked to do any of (1)–(5), respond with the appropriate handoff template from the "Approved Response Templates" section below — do not produce the number. Ambiguous-Term Clarification (before any analysis) -------------------------------------------------- If the user's question uses an ambiguous term, ask before computing anything. Examples: - **"aggressive"** → could mean shorter overall survival, higher metastasis rate, higher grade/stage, higher TMB, or specific molecular features. Ask which definit … (9984 more chars) ▶ list_studies { "search": "glioblastoma" } ◀ result {"result":[{"cancer_study_identifier":"gbm_tcga","name":"Glioblastoma Multiforme (TCGA, Firehose Legacy)","type_of_cancer_id":"difg","sample_count":619,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_tcga"},{"cancer_study_identifier":"gbm_tcga_gdc","name":"Glioblastoma Multiforme (TCGA GDC, 2025)","type_of_cancer_id":"difg","sample_count":611,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_tcga_gdc"},{"cancer_study_identifier":"gbm_tcga_pan_can_atlas_2018","name":"Glioblastoma Multiforme (TCGA, PanCancer Atlas)","type_of_cancer_id":"difg","sample_count":592,"has_guide":true,"url":"https://www.cbioportal.org/study/summary?id=gbm_tcga_pan_can_atlas_2018"},{"cancer_study_identifier":"gbm_tcga_pub2013","name":"Glioblastoma (TCGA, Cell 2013)","type_of_cancer_id":"difg","sample_count":577,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_tcga_pub2013"},{"cancer_study_identifier":"gbm_tcga_pub","name":"Glioblastoma (TCGA, Nature 2008)","type_of_cancer_id":"difg","sample_count":206,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_tcga_pub"},{"cancer_study_identifier":"gbm_cptac_2021","name":"Glioblastoma (CPTAC, Cell 2021)","type_of_cancer_id":"difg","sample_count":99,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_cptac_2021"},{"cancer_study_identifier":"gbm_columbia_2019","name":"Glioblastoma (Columbia, Nat Med. 2019)","type_of_cancer_id":"difg","sample_count":42,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_columbia_2019"},{"cancer_study_identifier":"gbm_iatlas_prins_2019","name":"Glioblastoma (Randomized Clinical Trial, Nat Med. 2019) - iAtlas Harmonized","type_of_cancer_id":"adifg","sample_count":30,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=gbm_iatlas_prins_2019"}]} ═ answer (success) Several glioblastoma studies are available, and the counts would differ by study. Which one do you want? - [Glioblastoma Multiforme (TCGA, PanCancer Atlas)](https://www.cbioportal.org/study/summary?id=gbm_tcga_pan_can_atlas_2018) (`gbm_tcga_pan_can_atlas_2018`, 592 samples) - [Glioblastoma Multiforme (TCGA, Firehose Legacy)](https://www.cbioportal.org/study/summary?id=gbm_tcga) (`gbm_tcga`, 619 samples) - [Glioblastoma Multiforme (TCGA GDC, 2025)](https://www.cbioportal.org/study/summary?id=gbm_tcga_gdc) (`gbm_tcga_gdc`, 611 samples) - [Glioblastoma (TCGA, Cell 2013)](https://www.cbioportal.org/study/summary?id=gbm_tcga_pub2013) (`gbm_tcga_pub2013`, 577 samples) - [Glioblastoma (TCGA, Nature 2008)](https://www.cbioportal.org/study/summary?id=gbm_tcga_pub) (`gbm_tcga_pub`, 206 samples) - [Glioblastoma (CPTAC, Cell 2021)](https://www.cbioportal.org/study/summary?id=gbm_cptac_2021) (`gbm_cptac_2021`, 99 samples) - [Glioblastoma (Columbia, Nat Med. 2019)](https://www.cbioportal.org/study/summary?id=gbm_columbia_2019) (`gbm_columbia_2019`, 42 samples) - [Glioblastoma (Randomized Clinical Trial, Nat Med. 2019), iAtlas](https://www.cbioportal.org/study/summary?id=gbm_iatlas_prins_2019) (`gbm_iatlas_prins_2019`, 30 samples) You can pick one, or I can run all of them. The TCGA studies overlap heavily, so combining them would double-count patients. Two things to know about the result: - **Mutations only or copy number too?** These genes, especially CDKN2A, are often lost by homozygous deletion in glioblastoma rather than mutated. Tell me if you want mutations only or mutations plus deep deletions. - **What I can provide:** I can give you the counts of patients with each combination of alterations, using profiled patients as the denominator. I can't compute the statistical test for mutual exclusivity. The cBioPortal Mutual Exclusivity tab runs it for you, and I'll link to it for the study you choose.