Q (haiku): calculate median expression of ceacam5, itgb6, muc2, tpgb and muc1 mRNA in pancreatic cancer

▶ read_guide
{
 "uri": "cbioportal://gene-expression-guide"
}

▶ search_oncotree
{
 "search_term": "pancreatic cancer"
}

▶ list_studies
{
 "search": "pancreatic",
 "limit": 20
}

◀ result
# Gene Expression Analysis Guide

This guide covers continuous-value genomic data: gene **expression**, **copy number** values, **methylation**, and related profile types. Mutation/CNA/SV *frequency* analysis lives in `cbioportal://mutation-frequency-guide`.

## Where this data lives

Continuous per-sample-per-gene values are stored in `genetic_alteration_derived`:

| Column | Description |
|---|---|
| `sample_unique_id` | `<study>_<sample_stable_id>` |
| `cancer_study_identifier` | study scope |
| `hugo_gene_symbol` | gene |
| `profile_type` | which assay/normalization (see below) |
| `alteration_value` | the actual value — stored as Nullable(String); cast with `toFloat64OrNull` |

`alteration_value` is a string because the same column hosts many different value scales. The `''` and `'NA'` sentinels mean "missing"; always filter them out and use `toFloat64OrNull(alteration_value) IS NOT NULL` for downstream math.

## Discovering profile types for a study

Different studies expose different profile types depending on what assays were run and how the data was normalized. Always check what a specific study supports before picking one:

```sql
SELECT DISTINCT profile_type
FROM genetic_alteration_derived
WHERE cancer_study_identifier = 'brca_metabric'
ORDER BY profile_type;
```

Common values across the public portal:

| Family | Profile types |
|---|---|
| mRNA expression | `rna_seq_v2_mrna`, `rna_seq_v2_mrna_median_Zscores`, `rna_seq_v2_mrna_median_all_sample_Zscores` (TCGA PanCancer Atlas), `mrna`, `mrna_median_Zscores`, `mrna_seq_v2_rsem`, `mrna_seq_v2_rsem_Zscores`, `mrna_seq_cpm`, `mrna_seq_fpkm`, `mrna_U133`, `mrna_outliers` |
| Copy number (continuous) | `cna`, `linear_CNA`, `log2CNA`, `cna_consensus`, `cna_rae`, `gistic` |
| Methylation | `methylation_hm27`, `methylation_hm450`, `methylation_epic`, `methylation_promoters_rrbs` |
| miRNA | `mirna`, `mirna_median_Zscores` |
| Protein | `protein_quantification`, `protein_level`, `RPPA` |

**Z-score vs raw choice.** When the user asks "is X correlated with Y", either works for Spearman (rank-based) — Pearson would care. Default to the non-Z-score variant if both exist, and call out which one in the response.

## Canonical recipe — Spearman correlation between two genes

```sql
SELECT *
FROM gene_pair_coexpression(
    study        = 'brca_metabric',
    gene_a       = 'TP53',
    gene_b       = 'MYC',
    profile_type = 'mrna'
);
```

Returns one row: `(gene_a, gene_b, profile_type, spearman_correlation, num_samples)`.

- `spearman_correlation` in [−1, 1]; `NULL` when fewer than 3 valid paired samples.
- Mirrors cbioportal-backend's `ClickhouseCoExpressionMapper.getCoExpressions`, simplified to a pair lookup (the backend computes one ref gene vs ALL other genes for the coexpression page; here the agent asks about a specific pair).

### Verified examples

| Study | gene_a | gene_b | profile_type | spearman | n |
|---|---|---|---|---|---|
| `brca_metabric` | TP53 | MYC | `mrna` | 0.118 | 1980 |
|  … (4931 more chars)

◀ result
{"result":[{"code":"IOPN","name":"Intraductal Oncocytic Papillary Neoplasm","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PACT > IOPN"},{"code":"IPMN","name":"Intraductal Papillary Mucinous Neoplasm","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PACT > IPMN"},{"code":"ITPN","name":"Intraductal Tubulopapillary Neoplasm","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PACT > ITPN"},{"code":"MCN","name":"Mucinous Cystic Neoplasm","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PACT > MCN"},{"code":"OSGCT","name":"Osteoclastic Giant Cell Tumor","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > UCP > OSGCT"},{"code":"PAAC","name":"Acinar Cell Carcinoma of the Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PAAC"},{"code":"PAAD","name":"Pancreatic Adenocarcinoma","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PAAD"},{"code":"PAASC","name":"Adenosquamous Carcinoma of the Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PAASC"},{"code":"PACT","name":"Cystic Tumor of the Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PACT"},{"code":"PANCREAS","name":"Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS"},{"code":"PANEC","name":"Pancreatic Neuroendocrine Carcinoma","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PANEC"},{"code":"PANET","name":"Pancreatic Neuroendocrine Tumor","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PANET"},{"code":"PB","name":"Pancreatoblastoma","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PB"},{"code":"PSC","name":"Serous Cystadenoma of the Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > PACT > PSC"},{"code":"SPN","name":"Solid Pseudopapillary Neoplasm of the Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > SPN"},{"code":"UCP","name":"Undifferentiated Carcinoma of the Pancreas","score":65,"mainType":"Pancreatic Cancer","tissue":"Pancreas","hierarchy":"PANCREAS > UCP"}]}

◀ result
{"result":[{"cancer_study_identifier":"msk_met_2021","name":"MSK MetTropism (MSK, Cell 2021)","type_of_cancer_id":"mixed","sample_count":25775,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=msk_met_2021"},{"cancer_study_identifier":"pdac_msk_2024","name":"Pancreatic Adenocarcinoma (MSK, Nat Med 2024)","type_of_cancer_id":"pancreas","sample_count":2336,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=pdac_msk_2024"},{"cancer_study_identifier":"paad_qcmg_uq_2016","name":"Pancreatic Adenocarcinoma (QCMG, Nature 2016)","type_of_cancer_id":"paad","sample_count":456,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_qcmg_uq_2016"},{"cancer_study_identifier":"pancreas_ctdna_msk_2025","name":"Pancreatic Cancer cfDNA (MSK, J Natl Cancer Inst 2025)","type_of_cancer_id":"paad","sample_count":412,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=pancreas_ctdna_msk_2025"},{"cancer_study_identifier":"pancreas_msk_2024","name":"Pancreatic Cancer (MSK, Cancer Cell 2024)","type_of_cancer_id":"pancreas","sample_count":395,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=pancreas_msk_2024"},{"cancer_study_identifier":"paad_tcga_gdc","name":"Pancreatic Adenocarcinoma (TCGA GDC, 2025)","type_of_cancer_id":"paad","sample_count":186,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_tcga_gdc"},{"cancer_study_identifier":"paad_tcga","name":"Pancreatic Adenocarcinoma (TCGA, Firehose Legacy)","type_of_cancer_id":"paad","sample_count":186,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_tcga"},{"cancer_study_identifier":"paad_tcga_pan_can_atlas_2018","name":"Pancreatic Adenocarcinoma (TCGA, PanCancer Atlas)","type_of_cancer_id":"paad","sample_count":184,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_tcga_pan_can_atlas_2018"},{"cancer_study_identifier":"pancreas_cptac_gdc","name":"Pancreatic Cancer (CPTAC GDC, 2025)","type_of_cancer_id":"pancreas","sample_count":183,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=pancreas_cptac_gdc"},{"cancer_study_identifier":"paad_cptac_2021","name":"Pancreatic Ductal Adenocarcinoma (CPTAC, Cell 2021)","type_of_cancer_id":"paad","sample_count":140,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_cptac_2021"},{"cancer_study_identifier":"paad_utsw_2015","name":"Pancreatic Cancer (UTSW, Nat Commun 2015)","type_of_cancer_id":"paad","sample_count":109,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_utsw_2015"},{"cancer_study_identifier":"paad_icgc","name":"Pancreatic Adenocarcinoma (ICGC, Nature 2012)","type_of_cancer_id":"paad","sample_count":99,"has_guide":false,"url":"https://www.cbioportal.org/study/summary?id=paad_icgc"},{"cancer_study_identifier":"panet_arcnet_2017","name":"Pancreatic Neuroendocrine Tumors (Multi-Institute, Nature 2017)","type_of_cancer_id":"panet","sample_count":98,"has_gui … (1828 more chars)

▶ clickhouse_run_select_query
{
 "query": "
SELECT DISTINCT profile_type
FROM genetic_alteration_derived
WHERE cancer_study_identifier = 'pdac_msk_2024'
ORDER BY profile_type
LIMIT 50
"
}

◀ result
{"rows":[{"profile_type":"cna"}]}

▶ clickhouse_run_select_query
{
 "query": "
SELECT DISTINCT profile_type
FROM genetic_alteration_derived
WHERE cancer_study_identifier = 'paad_tcga_pan_can_atlas_2018'
ORDER BY profile_type
"
}

◀ result
{"rows":[{"profile_type":"gistic"},{"profile_type":"log2CNA"},{"profile_type":"rna_seq_v2_mrna"},{"profile_type":"rna_seq_v2_mrna_median_Zscores"},{"profile_type":"rna_seq_v2_mrna_median_all_sample_Zscores"},{"profile_type":"rppa"},{"profile_type":"rppa_Zscores"}]}

▶ clickhouse_run_select_query
{
 "query": "
SELECT hugo_gene_symbol,
       count() AS samples_with_data,
       round(median(toFloat64OrNull(alteration_value)), 2) AS median_rsem,
       round(quantile(0.25)(toFloat64OrNull(alteration_value)), 2) AS q25_rsem,
       round(quantile(0.75)(toFloat64OrNull(alteration_value)), 2) AS q75_rsem
FROM genetic_alteration_derived
WHERE cancer_study_identifier = 'paad_tcga_pan_can_atlas_2018'
  AND profile_type = 'rna_seq_v2_mrna'
  AND hugo_gene_symbol IN ('CEACAM5', 'ITGB6', 'MUC2', 'TPGB', 'MUC1')
  AND alteration_value NOT IN ('', 'NA')
  AND toFloat64OrNull(alteration_value) IS NOT NULL
GROUP BY hugo_gene_symbol
ORDER BY hugo_gene_symbol
",
 "max_rows": 10
}

◀ result
{"rows":[{"hugo_gene_symbol":"CEACAM5","samples_with_data":177,"median_rsem":10049.7,"q25_rsem":1115.14,"q75_rsem":32625.1},{"hugo_gene_symbol":"ITGB6","samples_with_data":177,"median_rsem":3740.15,"q25_rsem":2365.02,"q75_rsem":6166.43},{"hugo_gene_symbol":"MUC1","samples_with_data":177,"median_rsem":14799,"q25_rsem":7580.57,"q75_rsem":26641.5},{"hugo_gene_symbol":"MUC2","samples_with_data":177,"median_rsem":6.53,"q25_rsem":1.35,"q75_rsem":65.17}]}

═ answer (success)
## mRNA Expression Medians in Pancreatic Adenocarcinoma

**Study:** [Pancreatic Adenocarcinoma (TCGA, PanCancer Atlas)](https://www.cbioportal.org/study/summary?id=paad_tcga_pan_can_atlas_2018) — 184 samples, 177 with expression data

| Gene | Median RSEM | Q1–Q3 (IQR) |
|---|---|---|
| **CEACAM5** | 10,049.7 | 1,115–32,625 |
| **ITGB6** | 3,740.2 | 2,365–6,166 |
| **MUC1** | 14,799.0 | 7,581–26,642 |
| **MUC2** | 6.5 | 1.4–65.2 |
| **TPGB** | *Not found* | — |

**Data type:** RSEM (RNA-Seq by Expectation Maximization), batch-normalized across TCGA PanCancer Atlas studies for cross-cancer comparability. Raw counts, not Z-scores — values reflect absolute expression level.

**Note:** TPGB was not detected in the expression data for this study. If you're looking for an alternative gene name or want to check a different study (e.g., the larger MSK 2024 cohort has 2,336 samples but lacks mRNA data), let me know.

**Interpretation:** MUC1 shows the highest median expression; ITGB6 is moderately expressed across samples; CEACAM5 has high median but very wide spread (IQR spanning 30-fold). MUC2 is very lowly expressed, consistent with adenocarcinoma biology (not columnar/mucinous).