Our platform is enabled by our ability to efficiently analyze high-throughput molecular-level biochemical assays, including transcriptomics, genomics and/or proteomics, collectively referred to as Omics data. These different types of biochemical assays each provide us with unique information about the molecular mechanisms of disease biology and drug response. Since our inception, we have partnered with industry-leading pharmaceutical and biotechnology companies to perform a variety of analyses that utilize our expertise in translational bioinformatics. Examples publicly disclosed by our partners include our analyses of ibrutinib, ipilimumab, daratumumab, glatiramer acetate and pridopidine.

Underlying each of these elements is our rigorous quality control and ability to analyze complex biological datasets. We are one of the few oncology companies that has been involved in defining best practices for robustly analyzing bioinformatics data, as evidenced by co-authorship on journal articles together with regulatory experts as well as writing invited reviews to educate the scientific community on this topic. This attention to rigorous quality control pervades all of our analyses, and we believe this enables us to extract meaningful information from our own data as well as from a variety of databases of human data.


In early 2018, we began applying our proprietary platform and approach to internally develop our wholly owned pipeline of orally administered small molecule drug programs. Key elements of our platform include:

  • Insights from Human Data. Compare distinct groups of individuals who differ in a certain aspect of disease or response to a particular therapy, or identify new patient subsets. Our platform has enabled us to conduct multiple projects that involve stratifying patients into novel subsets. We associate transcriptomic profiles with each subset, which can then be directly inputted into DCT to identify novel targets specific to a given patient subset.
  • Novel Biology. Identify novel targets and new ways to drug existing targets using our Disease Cancelling Technology and/or our insights into mechanisms of response. Additional biologic context is derived from quantifying the extent to which different time points, concentrations and perturbations (e.g., inhibition and overexpression) may cancel a disease signal more effectively than existing drug targets.
  • Novel Chemistry. Rapidly identify small molecules that selectively bind to a target of interest using our proprietary Fluency deep learning AI technology, and/or engineer PK to achieve optimal signaling dynamics. Fluency identifies the most attractive drug candidates within a library by making ranked predictions of binding affinity for all compounds, and can be run on any library containing millions of compounds.
  • Proprietary Translational Planning. Use humanized preclinical models and bioinformatics to prioritize indications and identify sensitive subpopulations. In oncology, we are deeply experienced in advanced, humanized 3D-based tumor growth models, which based on peer reviewed research by members of our team and others, more accurately predict drug response in animal models, and we believe in patients, compared to standard models. Unlike in vitro approaches, the 3D tumor growth models reflect the complexity of tumor biology given their alignment with the TME.

Underlying each of these elements is our rigorous quality control and ability to analyze complex biological datasets.

Our approach played a critical role in determining the most important characteristics for and creation of our clinical stage product candidates, IMM-1-104 and IMM-6-415. Specifically, our platform enables us to:

  • Leverage insights from human data to identify disease transcriptional profiles we aim to counteract;
  • Identify novel biology, specifically evaluating new ways to drug an existing target by utilizing our proprietary Disease Canceling Technology, or DCT, and analyze mechanisms of existing drugs;
  • Generate novel chemistry that overcomes MAPK-feedback loops to achieve optimal signaling dynamics; and
  • Profile IMM-1-104 and IMM-6-415 in a large number of 3D models using our own translational planning to identify the types of cancer most likely to be sensitive to the product candidate.