Research Areas

Exploring the genomic landscape of cancer through computational approaches

Our Approach

The Georgakopoulos-Soares Lab combines computational biology, genomics, and machine learning to understand the complex patterns of mutations that drive cancer development and progression.

We develop innovative computational methods to analyze large-scale genomic datasets, with the goal of identifying new biomarkers, therapeutic targets, and insights into cancer biology.

Our interdisciplinary approach brings together expertise from computer science, statistics, molecular biology, and clinical oncology to address key challenges in cancer research.

Genomic Analysis

Analyzing cancer genomes to identify mutational patterns and signatures.

Machine Learning

Developing algorithms that surface actionable insights from complex data.

Translational Research

Bridging basic science discoveries with clinical applications.

Tool Development

Creating open resources for the scientific and medical community.

Research Focus Areas

Mutational Signatures in Cancer

Our lab is at the forefront of identifying and characterizing mutational signatures in cancer genomes. These signatures reveal the biological processes operating during tumor development.

By analyzing large-scale genomic datasets we discover new signatures and explore their biological origins with implications for etiology, detection, and treatment.

Key Projects:

  • Identification of novel mutational signatures in pediatric cancers
  • Computational methods for signature extraction from whole-genome sequencing
  • Linking mutational signatures to clinical outcomes
  • Characterization of tissue-specific mutational processes
Genomic AnalysisPattern RecognitionCancer Etiology

Genomic Instability and Cancer Evolution

Genomic instability drives tumor evolution and treatment resistance. We interrogate its mechanisms and consequences across cancer types.

Our computational methods quantify chromosomal instability, microsatellite instability, and replication stress to uncover therapeutic vulnerabilities.

Key Projects:

  • Cataloguing genomic instability patterns across tumors
  • Relating genomic instability to immunotherapy response
  • Tracking tumor evolution via instability markers
  • Identifying synthetic lethal partners of instability pathways
Tumor EvolutionGenomic InstabilityTreatment Resistance

Computational Methods for Cancer Genomics

We build computational and machine learning frameworks tailored to the complexity of cancer genomes.

Our tools integrate heterogeneous data, identify driver mutations, and deliver accessible software for the community.

Key Projects:

  • Multi-omics integration with machine learning
  • Frameworks for mutational signature analysis
  • Algorithms for discovering cancer driver mutations
  • Deep learning models for cancer classification and prognosis
Machine LearningAlgorithm DevelopmentData Integration

Translational Cancer Genomics

We translate genomic discoveries into clinical impact through biomarker development and validation.

Collaboration with clinical partners ensures our findings inform diagnosis, prognosis, and therapy selection.

Key Projects:

  • Biomarkers for early cancer detection
  • Predictive markers for immunotherapy response
  • Prognostic models based on mutational signatures
  • ctDNA monitoring for treatment response
Biomarker DiscoveryPrecision MedicineClinical Translation

Our Methods & Technologies

Computational Genomics

Advanced pipelines for whole-genome, whole-exome, and transcriptomic data analysis.

Machine Learning

Supervised, unsupervised, and deep learning models tailored to genomic signals.

Data Integration

Combining genomic, epigenomic, transcriptomic, and clinical data for holistic insights.

Statistical Modeling

Robust statistical frameworks for mutational processes and clinical associations.

Software & Resources

SignatureExplorer

A comprehensive suite for extracting and interpreting mutational signatures.

GitHubDocumentation

GenomeInstability

Toolkit for quantifying genomic instability across patient cohorts.

GitHubDocumentation

MultiOmicsIntegrator

Framework for harmonizing multi-omics data and generating integrative models.

GitHub