Researchers at the Harvard T.H. Chan School of Public Health have created a powerful new computational workflow that could help scientists make better use of one of the world’s largest cancer data resources, The Cancer Genome Atlas (TCGA). TCGA contains massive amounts of genomic information, including RNA sequencing, DNA methylation, and other molecular data from thousands of patient samples across many cancer types. These datasets hold clues about how genes are switched on or off in cancer, how tumors evolve, and why some respond differently to treatment. However, analyzing TCGA data has traditionally required researchers to use multiple software tools, each with its own setup and compatibility issues.

To address this challenge, the team developed tcga-data-nf, an open-source, Nextflow-based workflow that brings everything together into a single, easy-to-use pipeline. With one command, users can download TCGA data, clean and pre-process it, and then use computational models to build gene regulatory networks, maps that show how genes and proteins interact to control cellular behavior. The workflow integrates multi-omic data, including RNA sequencing and methylation information, to help researchers explore how changes in gene regulation contribute to cancer development and progression.

A key feature of tcga-data-nf is its reproducibility. In bioinformatics, reproducibility ensures that an analysis can be repeated by different researchers and yield the same results, which is critical for scientific validation. tcga-data-nf standardizes every step of the data processing and analysis, minimizing inconsistencies that often arise when different methods or software versions are used.

The researchers also developed an accompanying R package called NetworkDataCompanion, which manages and filters TCGA data for network modeling. Using this integrated framework, they demonstrated how the workflow can be applied to study colon cancer subtypes, revealing differences in gene regulation linked to epigenetic mechanisms, changes that affect gene activity without altering DNA sequences.

Beyond the workflow itself, the team compiled a database of pre-generated regulatory networks for the ten most common cancer types in TCGA. This resource allows researchers to explore and compare gene interactions across cancers without having to run complex analyses from scratch.

By unifying existing tools into a cohesive, automated system, tcga-data-nf lowers the barrier to entry for cancer researchers who want to analyze large-scale genomic data. It helps scientists generate biologically meaningful insights more efficiently, supports transparency in data-driven research, and opens the door to new discoveries about the molecular underpinnings of cancer.

Availability – https://github.com/QuackenbushLab/tcga-data-nf