# Gene Activation (ANOVA) 2025 — Data

来源：https://www.kaggle.com/competitions/gene-activation-anova-2025/data

采集时间：2026-10-20T03:19:48Z

The dataset includes 1,000 individuals and 500 genes. Each individual has:

* A measured wellbeing score,
* Binary flags indicating whether a gene is expressed,
* Demographic characteristics commonly associated with health outcomes.

This dataset is observational and contains correlated predictors. Some genes are redundant, some are confounded with demographic variables, and only a small subset is truly causal.

No ground truth gene effects are provided.

---

## Files

The competition provides the following files:

* `gen_data.csv`
* `sample_submission.csv`

---

## Files Columns

gen_data.csv

Each row corresponds to one individual.

Columns:

* `wellbeing_index` — The response variable
* `Gene_1` … `Gene_500` — Binary gene activation indicators (0 = inactive, 1 = active)
* `sex` — Sex indicator
* `age` — Age in years
* `height` — Height in meters
* `weight` — Weight in kilos

---

sample_submission.csv

Columns:

* `colID` — Gene identifier (Gene_1, Gene_2, …)
* `Probability` — Predicted probability that the gene’s intervention increases wellbeing by at least one point