Multi-omic Assessment of Squamous cell cancers receiving Systemic Therapy (MASST)

Identifying the small number of patients who respond to immunotherapy and predicting survival using traditional biomarker testing is difficult. MASST is developing a computational platform and methods to jointly analyze multimodal molecular, imaging and clinical data to identify patients that would benefit most from immunotherapy.


The 2BLAST Pipeline Development: (Biostatistical and Bioinformatic Longitudinal Analysis of Symptoms and Toxicities in cancer patients)

Understanding patterns of symptoms and toxicity data for cancer patients is essential for improving symptom, therapeutic and patient management. Leveraging this data is challenging due to sheer volume and complexity: it is high-dimensional, correlated, messy and changes through time. There are no existing pipelines for integrating symptoms and toxicity data with detailed clinical data, for research or clinical practice. 2BLAST is enabling routine automated processes for integration, quality assurance and analysis of institution-wide data to help clinical patient management and support a wide range of research endeavors from genomics to health services research.


MIRA Clinical Learning Environment (MIRA-CLE) for Lung

Translation of data science research into clinical practice requires automated pipelines tied into the entire treatment pathway to facilitate real-time analysis and decision support. MIRACLE will help deliver patient-centered, value-based healthcare by identifying lung cancer patients who may be at higher risk of (1) inflammatory lung disease (ILD), (2) local failures, distant metastases and reduced overall survival based on tumour specific growth rate (SGR) and (3) treatment toxicities based on comparison of cone beam CT (CBCT) images over time. Identifying these patients pre-emptively enables better patient selection and individualized care.


To learn more about our Call for Catalyst Projects, please click here!



The PM Data Science Program is participating in a collaboration with MAASTRO, a renowned radiotherapy centre in the Netherlands, to establish an international, federated learning network for prognostics using open-source data.


Screenshot of Biospecimen Portal

BioSpecimen Portal

Biospecimen Portal is a web-based tool developed for the Research community at UHN by the PM Data Science Development team at TECHNA Institute. Governance for Biospecimen Portal falls under the Princess Margaret Data Science program. The tool allows researchers to search a subset of the caTissue database to find biospecimens related to those currently in use in their labs.

Screenshot for UHN model tracker

UHN Tumour Model Tracker

Tumour Model Tracker is a web-based tool developed for the Research community at UHN by the PM Data Science Development Team under the governance of the Princess Margaret Data Science Program. The tool allows researchers to search for tumour models, studies producing tumour models and short tandem repeat (STR) results for both patient- and model-derived samples.

Princess Margaret Integrated Dashboard

One of the objectives of the Princess Margaret Data Science Program is to increase access to data for the broader PM community in a way that will help optimize research and clinical efforts. The COVID-19 pandemic has accelerated our roadmap for this initiative.

Research Data Storage (RDS)

The PM Data Science Program is undertaking a CFI-funded initiative to centralize research data and increase data findability through the Research Data Storage (RDS) project. There is a large amount of data generated by research equipment across UHN. The data is coming from a variety of equipment, such as confocal microscopes, flow cytometers, and mass spectrometers, as well as a variety of labs and research initiatives. In most cases, the data’s value is limited to the project for which it was collected, and the possibilities for its reuse are limited. A centralized and standardized storage solution is needed to allow for a systematic approach to data management.


MIRA is a software framework developed by Tony Tadic and Tirth Patel, alongside a multidisciplinary team within the Radiation Medicine Program (RMP) and the PM Data Science program at UHN. MIRA enables the development and deployment of imaging pipelines, prediction models, and machine learning algorithms to support Data Science and machine intelligence at UHN.


PMDS provides support to essential software for oncology research at PM.
The following software applications are currently being supported: 


A software tool to mine and analyze multiple large cancer pharmacogenomic datasets

A software tool to link clinical and genomic data from patients with cancer across the country


Software for segmentation and genome annotation.