Glioma Grading Clinical and Mutation Features from TCGA Projects

Patient records from The Cancer Genome Atlas (TCGA) LGG and GBM projects, characterized by 20 gene mutation statuses and 3 clinical features. The dataset was created for the prediction task of classifying glioma grade as Lower-Grade Glioma or Glioblastoma Multiforme. It was funded by The Cancer Genome Atlas (TCGA) Project - NCI and authored by Tasci, E., Zhuge, Y., Kaur, H., Camphausen, K., & Krauze, A. V.

Use Cases

• Build a binary classifier for glioma grading (LGG vs. GBM) based on clinical and molecular features.
• Perform feature selection to identify an optimal subset of the 20 mutation genes and 3 clinical features to reduce diagnostic costs.
• Analyze the relationship between specific gene mutations and glioma tumor grade.
• Validate machine learning models for medical diagnosis using 10-fold cross-validation as suggested in the description.

Strengths

• Includes 20 molecular features and 3 clinical features per patient record, providing a multi-faceted view for classification.
• Data preprocessing removed 23 instances with missing values in Gender, Age_at_diagnosis, or Race, likely improving data consistency.
• Derived from authoritative TCGA-LGG and TCGA-GBM projects, ensuring a clinically relevant foundation.

Limitations

• Row count is unknown, which may limit suitability assessment for model training.
• Column-level documentation is absent; field semantics for the 20 genes must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

The Cancer Genome Atlas (TCGA) Project - NCI

Collection Method

Constructed from TCGA-LGG and TCGA-GBM brain glioma projects.

Time Range

null

Freshness

null

Geography

null