Unleashing the power of single-cell technologies

What are the advantages of targeting the whole cell rather than a single molecular target in drug creation?

Cellarity’s platform is an entirely new approach to drug discovery. Focusing on the cell allows us to see biological systems in full, and not through the proxy of a single target. Our AI models utilise data at single cell resolution to identify cell state transitions that drive disease and to identify compounds that can revert these disease state transitions. Our approach allows us to uncover novel actionable biology, even in diseases with no known targets, and design non-intuitive chemistry. Also, since the cell is a much fuller representation of disease, our platform is designed to drive much higher clinical success.

How does the drug creation approach that targets the whole cell help to unravel the complexity of disease biology?

We apply our cell centric platform to discover novel small molecules that target the cellular mechanisms of disease. For instance, for many chronic diseases, which are the leading cause of disability and health care costs in the US, we do not know the underlying molecular pathology. Many of these complex diseases are not driven by a single target or genetic driver and have therefore defied our current drug discovery paradigms. In contrast, because they are driven by complex regulatory changes, they are a great fit for the Cellarity platform. (figure 1)

What are some of the key challenges and limitations of targeting the whole cell in drug creation, and how are these being addressed?

Previously, the ability to target the whole cell was limited as the technology did not exist, but with the emergence of new technologies like single cell sequencing, we can now utilise the whole cell to uncover novel biology. Since this technology is so new, as is our ability to analyse the large amount of data volumes with computation, this means we are creating the data and analytical methods from scratch, which is one of the key challenges we face in the continued development of our platform.

How does AI/ML help in the analysis of single-cell technologies to better understand the transition of cells from health to disease?

Without AI/ML, our ability to learn from the vast amount of data that single-cell technology produces would be severely limited. At Cellarity, our platform utilises proprietary AI models trained on over 30-million single cell transcriptomes allowing us to uncover novel biology. The AI allows us to see patterns in the data that we would not have otherwise been able to see and provides an unbiased analysis, all of which helps us create non-intuitive drug candidates in a vast array of diseases.

What challenges and limitations do you see in applying AI/ML to the analysis of single-cell technologies for drug discovery?

AI is not necessarily the limitation but rather the data inputs that are required. Depending on the biological questions that need to be addressed, sourcing high quality samples, and generating and analysing the right data types is essential.

What are some examples of diseases that have no known targets but can be studied using this new approach to drug discovery?

There are many diseases, such as NASH, that have multiple known targets but no successful treatments. For other diseases, there are known targets, but they are not druggable. Cellarity’s approach can be applied to virtually any disease linked to dysfunctional cellular biology. The cell is a much fuller representation of disease, allowing us to see biological systems more holistically and not through the proxy of a single target or pathway and therefore, we can identify novel actionable biology even in areas where there is a lack of known targets.

How can this approach to drug discovery potentially lead to the development of more effective treatments for diseases?

In sickle-cell disease, we generated a single-cell map of haematopoiesis and identified cell behaviors tied to the production of a protective form of haemoglobin. We then used our AI systems to predict small molecules that directly induce these cell behaviors. These compounds achieved remarkable results, with efficacy exceeding standard of care and equivalent to gene therapy in vitro. Examples like this show how our platform can revolutionise the treatment of sickle-cell disease and other conditions by addressing the root causes of disease at the cellular level.

What is Cellarity doing to advance this field?

As previously mentioned, our platform is fueled by new technologies and analytical methods which had not previously been developed. Knowing there is still much for the community to learn, we have now twice released novel, open-sourced, single-cell datasets and invited the ML communities to develop new algorithms that can help the community at-large learn the rules of cell behavior. Also, last year, we organised the first Single Cell and AI in Medicine Symposium, and in May 2023, hosted it for a second time, because we saw the need for a forum for scientists to discuss the state of innovation at the intersection of single-cell analysis, AI, and biology and how this field is impacting drug discovery and the progress of medical science.