How to analyse tumor data and publish the results

Background

You need to process tumor sequencing data to extract key genomic alterations. Cancer genomics portals help you explore and share these results, but they require data in a specific structured format.

Scenario Overview

You start with raw tumor sequencing files, run an analysis workflow to generate variants and copy number profiles, then format the outputs so they can be uploaded to a cancer genomics portal for exploration.

Complexity

High

Environment

HPC or Cloud

Outcome

Published dataset

Your Journey

Set up HPC environment

Ensure you have access to computational infrastructure.

You have access to an HPC cluster or cloud environment with sufficient compute resources and storage to run the analysis pipeline. The environment should have Nextflow and container support (Singularity/Docker) installed.

Examples

Bioinformatician

I've requested access to our institutional HPC cluster and confirmed I can submit jobs with the required memory and CPU resources for whole genome analysis.

Research Engineer

I've set up a cloud compute environment with the necessary quotas and installed Nextflow to orchestrate the pipeline execution.

Collect and stage tumor sequencing data

Ensure sequencing files are accessible from the compute environment.

You've obtained the tumor and normal sequencing data files (FASTQ or BAM format) from your sequencing facility and made them accessible from your HPC or cloud storage. The data should be organized and reachable from where you'll run the analysis pipeline.

Examples

Clinical Researcher

Our tumor samples have been sequenced and I've transferred the raw files to our HPC shared storage where the pipeline can access them.

Bioinformatician

I've downloaded the sequencing data from the sequencing facility's server to our HPC scratch space and organized the tumor-normal pairs for processing.

Run the tumor analysis

Process tumor sequencing data to generate variants and profiles.

Use oncoanalyser to process the tumor sequencing data and generate variants, copy number profiles and other outputs.

Tools & Resources

Oncoanalyser View tool details

Examples

Bioinformatician

I need to process whole genome sequencing data from a tumor sample to identify somatic variants, copy number alterations, and structural variants for downstream analysis.

Clinical Researcher

Our lab has tumor samples sequenced and I need to run the analysis pipeline to extract the genomic alterations that might be clinically actionable.

Prepare the portal-ready dataset

Convert pipeline outputs to portal-compatible formats.

Take the standard outputs from oncoanalyser (VCF, copy number tables, fusion calls, expression files) and convert them into the simple tabular formats and study metadata structure required by cancer genomics portals. This step connects the pipeline results to the portal by mapping each output type to its corresponding portal file.

Tools & Resources

oncoanalyser outputs Documentation

cBioPortal file formats Documentation

Examples

Bioinformatician

I have the oncoanalyser VCF and copy number outputs and need to reformat them into the specific MAF and SEG files that cBioPortal expects for data import.

Data Manager

I need to ensure the clinical annotations and sample metadata are properly structured to match the portal's data model before uploading.

Load the dataset into cBioPortal

Upload your prepared data to the portal.

Use the cBioPortal import process to upload the prepared study directory to make your tumor data available in the portal for exploration.

Tools & Resources

cBioPortal View tool details

Examples

Data Manager

I need to import our prepared study into our institutional cBioPortal instance using the data loading scripts.

Bioinformatician

I'm uploading the formatted datasets to cBioPortal so the research team can start exploring the results.

Analysed data published in portal

Success! Your data is now accessible for exploration.

Congratulations! Your tumor genomics data is now published in cBioPortal. Researchers can interactively explore mutations, copy number alterations, gene expression, and clinical associations through the portal's visualization tools.

Examples

Cancer Researcher

Our study data is live in cBioPortal and the team can now query mutation patterns, survival associations, and genomic correlations without needing bioinformatics support.

Clinical Collaborator

I can now share the portal link with clinical partners who can explore our findings visually and identify potentially actionable alterations.