Harnessing the power of AI technology for drug discovery treatments

The talk of the power of artificial intelligence (AI) to either save or destroy humanity is pretty deafening. One way of explaining what people mean by AI is that it encompasses the technologies we use to let computers sift through large data sets with programs (“algorithms”) to find and extract patterns to help humans do this to understand these patterns, to suggest decisions about future decisions and to predict future events or suggest outcomes not previously foreseen.

Although some expect AI to mimic human intelligence, some AI systems show performance superior to humans (e.g. AlphaGo1). On the other hand, unlike humans, most AI systems require considerable data and training to achieve acceptable performance (e.g. we only have to taste strawberries once to know whether we like them or not).

One area of ​​research that has attracted a lot of interest and funding is the AI-based design of novel drugs – particularly small molecules.2 A simple way to classify the work done in this area is shown in the figure.

focus on one goal

Goal-based approaches are based on a a priori Selection of a receptor or enzyme to be activated, inhibited or modulated.3 For example, targets for psychiatry could be monoamine transporters (e.g. SSRIs) or newer targets like TAAR1 (e.g. ulotaront4). A receptor model implies knowledge or hypothesis of its structure based on known crystalline structures.5 A ligand-based model requires a data set that captures the structure of target-selective molecules (“ligands” and “chemical probes”) and the outcome of their interaction with a target protein (e.g., a neurotransmitter receptor) in an appropriate assay (e.g., .the serotonin 1A receptor6).

Importantly, both receptor- and ligand-based approaches for New Drug design implicitly assumes that action at that particular target is the most important effect that determines a drug’s therapeutic value.

The advantage of these modeling platforms is that existing data can be used in silico to predict the effect of a new molecule on a target, as well as its physical properties, before triggering costly preclinical work. Since building AI expertise also involves significant costs, various partnerships between AI companies and pharmaceutical companies have been initiated to advance the development of AI-based models New Drug design, avoiding the build-up of in-house AI expertise.

Partnerships such as Genentech/GNS Healthcare, GSK/Insilico Medicine, Takeda/Numerate AI, Atomwise/Abbvie, CrystalGenomics/Standigm and Cloud Pharmaceuticals/TheraMetrics use novel machine learning techniques to design or identify molecules that act on biological targets of particular interest .7 However, most of these partnerships do not focus on psychiatric indications or on CNS disorders in general.

Phenotypic drug discovery: A holistic approach

Drug research for mental health, whether traditional or AI-based, is lagging behind, while mental health needs have continued to grow and remain a major societal burden. Although a single target-based AI drug design may be appropriate for some indications, evidence from psychiatric research points to the need for polypharmacology (Figure). Indeed, action may be required on different objectives aiming at an appropriate balance.8th

However, to quantify the effect of compounds at multiple targets in a way that includes downstream effects and interactions, it is necessary to study drug activity with one in vivo Phenotypic drug discovery (iPDD) approach.9 An iPDD platform can be used for polypharmacology against known multi-targets or in a target-agnostic manner (Figure), since the organism used for drug screening acts as an enhancer of the effects of all compounds, providing a comprehensive drug profile. iPDD platforms include these high-throughput screens based on Drosophila, zebrafish and mice.10-12

Phenotypic screening with an iPDD platform enables characterization of the full spectrum of behavioral effects of reference drugs and data-driven, targeted comparison with novel compounds. The potential of machine learning-based analysis of behavioral phenotyping data can be used in many drug discovery applications, e.g. B. searching iPDD-screened compound libraries in search of new hits or analyzing novel analogues of drug candidates to speed up the lengthy drug discovery process. In addition, the use of animal models of diseases in iPDD platforms opens up opportunities to explore phenotypic drug screening for psychiatric, neurodegenerative and orphan diseases.13

iPDD drug discovery projects can proceed agnostically or morph into multi- or single-target programs as needed. Although ignorance of the mechanism of action makes it difficult to get to the clinic, the risk of target-agnostic programs can be reduced by “anti-target” panels, e.g. B. by avoiding D2 antagonism in the development of novel antipsychotics. Biomarkers can also be used to assess target binding in the clinic when the mechanism of action is unknown.

Perhaps the most compelling evidence will be at hand very soon. Ulotrone, an antipsychotic with a novel mechanism of action now in Phase III testing, was discovered and developed in a collaboration between Sunovion and PsychoGenics,4 using an iPDD platform (SmartCube®).

Final Thoughts

In summary: target and ligand-based AI New Drug design approaches show promise for indications with validated hypotheses on the required therapeutic mechanisms of action. In contrast, complex CNS diseases require phenotypic screens and associated AI methods (e.g. iPDD platforms). The phenotypic approach in the coming era of AI-based drug design promises to bring new insights and accelerate drug discovery for CNS disorders.

Dr Bruner is Chief Innovation Officer at PsychoGenics Inc. and Associate Associate Professor at Mt. Sinai School of Medicine. She is a member of CTF’s Business Advisory Board and CureVCP’s Scientific Advisory Board.

references

1. AlphaGo. deep mind Retrieved October 12, 2022. https://www.deepmind.com/research/highlighted-research/alphago

2. Vedantam K. AI is making its way into drug research. What does this mean for biotechnology? crunch base. October 4, 2022. Accessed October 13, 2022. https://news.crunchbase.com/health-wellness-biotech/artificial-intelligence-venture-drug-discovery/

3. Mouchlis VD, Afantitis A, Serra A, et al. Advances in de novo drug design: from conventional to machine learning methods. international J. Mol. Sci. 2021;22(4):1676.

4. Correll CU, Koblan KS, Hopkins SC, et al. Safety and Efficacy of Ulotaronte (SEP-363856) in Schizophrenia: Results of a 6-month, open-label extension study. NPJ Schizophr. 2021;7(1):63.

5. Schwartz TW, Hubbell WL. Structural biology: a moving history of receptors. Nature. 2008;455(7212):473-474.

6. Czub N, Pacławski A, Szlęk J, Mendyk A. Do AutoML-based QSAR models comply with the OECD principles for regulatory assessment? A 5-HT1A receptor case. pharmacy. 2022;14(7):1415.

7. Buvailo A. How big pharmaceutical companies are using AI to advance drug discovery. BioPharmaTrend.com. October 8, 2018. Accessed October 12, 2022. https://www.biopharmatrend.com/post/34-biopharmas-hunt-for-artificial-intelligence-who-does-what/

8. Kondej M, Stepnicki P, Kaczor AA. Multi-target approach to drug discovery against schizophrenia. international J. Mol. Sci. 2018;19(10):3105.

9. Leahy E, Brunner D. We need a new Prozac: the demand for innovative brain drugs. The pharmaceutical letter. August 15, 2022. Accessed October 12, 2022. https://www.thepharmaletter.com/article/we-need-a-new-prozac-the-demand-for-brain-drug-innovation

10. So TT. drug screening in Drosophila; why, when and when not? Wiley Interdiscipline Rev Dev Biol. 2019;8(6):e346.

11. McCarroll MN, Gendelev L, Keizer MJ, Kokel D. Using large-scale behavioral profiling in zebrafish to explore neuroactive polypharmacology. ACS Chem. Biol. 2016;11(4):842-849.

12. Roberds SL, Filippov I, Alexandrov V, et al. Rapid, computer vision-enabled mouse screening system identifies neuropharmacological potential of two novel mechanisms. Front neurosci. 2011;5:103.

13. P. Kabitzke, D. Morales, D. He, et al. Mouse model systems of autism spectrum disorder: replicability and informatics signature. genes brain behavior. 2020;19(7):e12676.

source

Leave a Reply

Your email address will not be published. Required fields are marked *