OneThree Biotech was born out of the largest precision medicine institute in America with the purpose of decreasing the failure rate of bringing therapeutics to market using biology driven artificial intelligence. Our technology was builtin a lab setting; allowing us to engineer our methodologies around real-world results. While working with major biotech and pharma partners; we’ve validated our predictions through work in target discovery, toxicity prediction, biomarker selection, and much more. In this white paper, we’ll discuss various OneThree modules can be combined to create a novel approach to target and compound discovery in oncology.


Targeted therapies for cancer patients have proven to be effective and many times life saving treatments. However, there are still many patients that do not respond to any approved treatment or quickly acquire resistance, highlighting a clear need to discover novel oncology targets. Prospective new targets can be effective drug repurposing opportunities if they are discovered to already be targeted by a developed compound. However, identifying new targets requires extensive experimentation for any specific cancer subtype. Additionally, the full list of a compound’s targets is seldom known. To help address this problem we’ve applied multiple modules in the OneThree AI platform to provide a wide-lens approach to target and compound discovery.

OneThree's Workflow for Target and Compound Discovery

Target Discovery

Through the combined efforts of various OTB modules, researchers are capable of identifying novel oncology targets that are both safe and efficacious. First, novel targets, that will be sensitive to the inhibition, can be identified with the OTB Gene Essentiality module. This allows for researchers to identify novel targets for any cancer type, subtype, or patient cohort of interest. This specificity within predictions, enables more refined positioning efforts, such as resistant populations or genomically similar cohorts. Once we identify a list of possible targets for the cancer subtype of interest, these targets can then be investigated for likelihood of toxicity by utilizing the OTB Adverse Event module. Finally, a refined list of both efficacious and safe, context specific targets is generated and downstream compound development can begin. 

Compound Discovery

From a target of interest, the OTB modules can be utilized to identify matching compounds, with associated biomarkers, that are safe and efficacious. The OTB database can be used to match the identified target to any investigational, shelved, or approved compounds that are known to inhibit the target. If none of the compounds that are identified are of interest for any reason (commercial viability, binding affinity, etc.), we can utilize OTB’s Target Deconvolution module and predict what other compounds will inhibit our target of interest. Once a list of compounds is created, researchers can filter which molecules are likely to cause any type of adverse event, using the OTB Adverse Event module. The last step will be to leverage the OTB Biomarker Identification module to identify which specific patient populations will be most sensitive to our compound candidates. OTB’s various modules can work in harmony to ultimately identify strong drug repurposing candidates for oncology. 

Relevant Publications:

  1. A machine learning approach predicts essential genes and pharmacological targets in cancer
  2. A Bayesian machine learning approach for drug target identification using diverse data types