Computational Discovery
Computational Approach
Immuno-oncology represents a paradigm shift in the treatment of cancer, and biological drugs blocking immune checkpoint targets have already resulted in long-term patient survival in certain cancer types. Despite their potential, current checkpoint inhibitors are limited to a few targets and are only effective in 20-30% of patients and in certain cancers. We believe that the identification of new drug targets and new biological pathways has the potential to broaden the reach of cancer immunotherapies to more types of cancers and many more patients.
We discover novel immuno-oncology drug targets through a unique, predictive, cloud-based computational approach.
Our in-silico targets discovery platforms combine our expertise in data sciences as applied in the analysis of vast amounts of publicly available and proprietary data sets. Our multi-omics data analysis is designed to identify novel drug target candidates for the development of first in class drugs. We contend that biology, including drug targets discovery, consists of complex scientific systems requiring an integrative multi-dimensional approach relying on multiple tools to generate the best results, and preferably should not be bound to a specific technology. Our target discovery process is flexible, enabling tailor-made approaches designed to address unmet clinical needs. These consists of a toolbox of various omics data, a suite of computational solutions and purpose-built algorithms, augmented with human expertise to optimize and review the output.
As a result of our strategic computational approach, we have:
Developed a unique pipeline of drugs in development with the goal of expanding the number of patients that respond to cancer immunotherapy.
Entered into multiple strategic immuno-oncology collaborations with leading pharmaceutical companies and academic institutions.
Filed over 120 granted or pending patents and published over 85 peer reviewed publications.
Pioneering predictive computational discovery platform
From target discovery to clinical validation


Predictive Drug Discovery
We have developed predictive drug target discovery platforms that leverages the power of computational modeling, guided by our scientific expertise and extensive public and proprietary datasets, to identify novel drug targets and new biological pathways towards the development of new cancer immunotherapy treatments. We believe that our computational approach (now validated by three Compugen-discovered targets being evaluated in the clinic) integrated with robust experimental validation is a key differentiator from others employing computational discovery approaches.
Our broadly applicable predictive drug target discovery platforms employ a suite of cloud-based computational solutions and purpose-built algorithms to sort through both public and proprietary datasets encompassing genomics, transcriptomics, and proteomics data.
From these massive datasets, our platforms analyze characteristics, such as gene structure, protein domains, predicted cellular localization, expression pattern, as well as other characteristics to identify potential druggable targets and predict their biological functions.
Over the past decade, we have continued to refine our analysis by incorporating new public and in-house experimental data. While our initial focus is on discovering novel immune checkpoints, we are also working to identify myeloid targets contributing to the immunosuppressive tumor microenvironment as well as pathways driving drug resistance.
We have computationally identified multiple in-silico targets, including PVRIG, TIGIT and ILDR2, which now serve as the targets for therapeutic antibodies currently being evaluated in the clinic by us and others. Through our work and the work of others, these three targets are continually being validated as potential paradigm shifting discoveries in the field of cancer immunotherapy, supporting the capabilities of our discovery platforms and highlighting our ability to translate in-silico findings into experimentally validated clinical candidates.
Drug Target Candidate Selection
Our predictive target discovery process is followed by an experimental validation phase to evaluate the in-silico prediction and provide deep scientific understanding of the new drug targets and biological pathways that can be fed back into our computational platforms to further optimize our proprietary solutions. This process is performed in-house as well as via scientific collaborations with top academic laboratories. The most promising targets are advanced to antibody therapeutics pipeline, where we develop biologics capable of targeting the proteins of interest.
Biomarker Driven Strategy
At Compugen, we recognize that one of the major limitations of current immunotherapy approaches is the lack of tools to help us predict patient responses. Through the use of informed biomarker driven strategies, based on the new biological pathways we discover, we aim to identify biomarkers that can help us predict which patients are most likely to respond to our novel therapies. This long-term approach also seeks to improve the probability of success of our clinical studies.
Our clinical trials are designed to advance our biomarker driven strategy. Not only are we utilizing the extensive preclinical data we have collected on biomarker expression in individual tumor types to inform the study of our drugs and the indications that we are evaluating, but we are also collecting tissue samples from patients in order to create data characterizing tumor physiology before and during treatment to be utilized in future discoveries.