Ulcerative Colitis


Better target discovery and compound design in Ulcerative Colitis

What is UC?

Ulcerative colitis (UC) is a chronic, lifelong disease that causes inflammation and ulceration of the inner lining of the colon and rectum, accompanied by debilitating symptoms.

Our work

There is no cure for UC, current treatments have side effects and don’t work on all patients. Our work focuses on developing safer and more effective oral small molecule treatments.

Progress so far

For our most advanced UC programme, we identified an AI-generated target with novel mechanisms of action in UC, and rapidly delivered a candidate drug to IND/CTA-enabling studies.

Latest News | BenevolentAI nominates pre-clinical candidate for novel ulcerative colitis target

Read more

UC affects 0.4% of the US population.(1)

20-40% of moderate-severe patients do not respond to anti-TNF therapy, a key treatment approach.(2)

The only oral treatment option currently available has serious side effects.(3)

$8.1bn–$14.9bn annual direct and indirect costs estimated in the US alone.(4)

Our work

Advancing safer and more effective small-molecule treatments for Ulcerative Colitis

The search for a novel target starts by scouring the vast available body of literature, a task akin to searching for a needle in a haystack, which often results in bias towards targets and biology that are well-studied or lie within a particular scientist's expertise within a given disease. Our data-driven approach enables discovery without boundaries or bias, removing conventional silos attributed to specific disease or therapeutic areas. This enabled us to identify a novel target with no previous link to UC.

Target identification: an unbiased approach to uncovering novel targets

Our Biomedical Knowledge Graph integrates all of the world’s available scientific data on all diseases to give a full and unbiased overview of genetics, pathology, chemistry, biological context and experimental data. We then use powerful AI and machine learning tools to analyse these multiple complex datasets to infer novel connections between targets and disease. Our scientific team traversed this interconnected data space using our graph exploration tools and identified an entirely novel target never previously associated with UC.

Omics: Building a deeper understanding of biology

We train a further class of AI models on specific data derived from UC patient populations to find patterns in the data and increase target quality. These include multi-modal ‘omics data sets and clinical data that enable scientists to better understand the cellular mechanisms and pathways underlying disease. Using our ‘omics workflow, bioinformaticians identified a distinct novel relationship between ulcerative colitis mechanisms and target expression, a finding which played a critical role in enhancing drug discoverers’ confidence when triaging our novel target for progression.

Chemistry: Designing the right drug

We use AI Chemistry to design the right molecule in order to maximise our chances of success in the clinic. In this stage, we identify a candidate drug optimised for many diverse properties. We combine machine learning models, generative design and novel 3D methodology to computationally model each of these complex endpoints and then optimise the molecular structure to achieve the best possible multi-parameter profile, so we can identify better compounds, faster.

Precision medicine: Finding the right patient subgroups

For our most advanced UC programme, we identified an AI-generated target with novel mechanisms of action in UC, and rapidly delivered a candidate drug within 2 years from the programme’s inception. We are actively using patient-derived molecular descriptors to identify the patient subgroups to optimise trial design but also to identify efficacy biomarkers to further increase our probability of success.

What's next

"Our UC programme demonstrates BenevolentAI's unique ability to uncover novel targets and accelerate the development of de-risked drug candidates by integrating molecular design and precision medicine. The programme unites the best of human and machine intelligence to build confidence in the early discovery phase, with the aim of ultimately providing new therapeutic options for patients.”

Nikki Robas  —  VP Drug Discovery, BenevolentAI

(1) Ulcerative Colitis: Epidemiology Forecast to 2029 - GlobalData Report Store, 2021. [Accessed April 2021]

(2) Roda, G et al. Loss of Response to Anti-TNFs: Definition, Epidemiology, and Management. Clin Transl Gastroenterol. 2016 Jan; 7(1): e135. doi: 10.1038/ctg.2015.63

(3) U.S. Food & Drug Administration. Initial safety trial results find increased risk of serious heart-related problems and cancer with arthritis and ulcerative colitis medicine Xeljanz, Xeljanz XR (tofacitinib). 2021. Available from: [Accessed April 2021].

(4) Cohen, R.D et al. Systematic review: The costs of ulcerative colitis in Western countries. Aliment. Pharmacol. Ther. 2010, 31, 693–707.