The Benevolent Platform™ is designed to optimise the earliest and most critical phase of the drug discovery process — target identification and validation
We use machine learning to unlock the power of the vast and expanding biomedical data landscape — including ‘omics, experimental data, literature and biological systems — to generate new insights into the underlying causes of disease.
Machine learning models extract biomedical entities, such as genes, diseases, processes and cell types, and infer relationships that capture how these entities interact in a human system. These relationships are stored in our Knowledge Graph as a network of contextualised scientific facts — providing a proprietary integrated view of biomedical data.
IMPACT
Our comprehensive data strategy allows us to work on any disease, or even pivot to tackle urgent public health issues. In early 2020, BenevolentAI scientists used our Knowledge Graph and tools to uncover a COVID-19 treatment1, which is now approved by the US FDA.2
Our target identification models and tools help scientists to better understand complex disease biology, make data-driven decisions and select the right drug target from the outset.
Our machine learning systems identify novel insights and relationships to reveal new hypotheses and propose drug targets that have never been considered for a disease before. We have also built AI tools to enhance target assessment and enable scientists to select only the most promising targets to take into wet lab experiments.
Case study
Scientists used our powerful target identification tools and machine learning models to rapidly identify and validate a novel biological target with no prior reference in published literature or patents linking the gene to ulcerative colitis.
Studies show that drug programmes grounded in patient or human genetic data have a greater chance of success.3,4 We integrate disease traits, genetics and genomics into our Knowledge Graph and use multi-omics data to generate endotype-specific target predictions that maximise our chance of clinical success.
We also identify individual genes relating to disease phenotypes, or clusters of genes that account for disease symptoms in certain patients. Our multi-omics data-specific machine learning models layer in personalised targets to the predictions list.
IMPACT
GBM has defied conventional research efforts due to the complexity, diversity and rapid growth of GBM tumours. Our specialist team are applying precision medicine techniques to identify more effective therapeutic targets that modulate tumour generating GBM stem cells.
We can validate high-quality drug targets generated by the Benevolent Platform™ in our fully-equipped laboratories in Cambridge (UK), where we have cutting-edge technologies such as CRISPR, RNA seq and human iPSCs to evaluate target hypotheses in cell-based models.
The more we do, the more we learn; we use experimental insights to enrich our Knowledge Graph and enhance future target predictions.
DRUG PIPELINE
Through the combined capabilities of our AI drug discovery Platform, scientific expertise, and wet lab facilities, we have rapidly built a substantial drug portfolio.
(1) Richardson et al. Lancet 395, e30-e31 (2020)
(2) Lilly Press Release 11 May 2022
(3) Nelson et al Nat Genet 47, 856-860 (2015)
(4) King et al. PLoS Genet 15, e1008489 (2019)
The Benevolent Platform™ has generated every programme in our in-house pipeline, which spans from target discovery to the clinic, mixing potentially best-in-class and first-in-class drug candidates.