Given there is no open source software available for the types of analysis required, proprietary algorithms are required. For this, mechanisms to extract data based on the objective of analysis are required together with the selection of the necessary biological entities and features related to the objective of analysis.
From this, AI can assist the scientist with interrogating the data to clarify ambiguity or verify the relevance of entities, helping the scientists on a faster path towards drug discovery. For example, a project might utilize a next generation platform to predict the absorption, distribution, metabolism, excretion, and toxicity of new drug candidates far faster than any traditional laboratory testing could achieve.
Hence, AI has the potential to provide deep insights into the continuum from chemical structure to in vitro, in vivo, and clinical outcomes.
Insilico Medicine
An example of how technology can be used to advance drug discovery comes from the company Insilico Medicine, a Honk Kong based biotech enterprise. The start-up launched the results of their artificial intelligence powered drug discovery process called GENTRL in 2020.
GENTRL is an acronym for Generative Tensorial Reinforcement Learning) As described in the study, GENTRL was able to successfully identify six promising treatments for fibrosis in just 21 days, with one of the treatments found to be effective in treating mice with renal fibrosis (the thickening and scarring of connective tissue in the kidney).
Based on this, researchers concluded that GENTRL promises to accelerate the drug discovery process, potentially saving the pharmaceutical industry billions, bringing affordable drugs to market faster and save lives as a result.
The reason why time savings are important when it comes to finding new medicines is because time equates to money and there is an estimated $2.6-billion price tag of developing a new treatment.
With the new application, molecules “imagined” by the GENTRL techniques were rapidly synthesized, tested in enzymatic, cell-based fibrosis, metabolic stability, microsomal stability assays, and in mice. Insilico Medicine assessed the GENTRL system for drug discovery in cells and animals and was able to create six new molecules that could address diseases like fibrosis.
Lowering costs, saving time
Another reason for the high cost is because of the many false starts, which can occur at any stage of the development process through to clinical trial. Advances in artificial intelligence are modernizing several aspects of our lives, deploying personified knowledge and learning from the solutions it produces to address not only specific but also complex problems. This also applies to pharmaceutical development, and evidence suggests that artificial intelligence can improve the efficiency of the drug development process and collaboration of pharmaceutical industry giants with AI-powered drug discovery firms.
Nanotechnology
To advance the detection of molecules and smaller nanoparticles (which present a vector for drug delivery) computational neural network are being developed which are capable of learning how the structure of a nanoparticle influences the way the particle scatters light. Such methods can aid in tackling research problems and doing so in ways that are considerably faster than any existing method is capable of achieving. The goal of this type of research in the biotechnological arena is to find ways to develop custom-designed, multilayered nanoparticles with properties of use for a range of different applications. Potential drug insights can arise as the artificial intelligence studies the structure and behaviors of nanoparticles, especially the ways that light of different colors are scattered light. As insights are gained about the relationships between light, color and structure, the platform can be run backwards to propose novel design for nanoparticles (this is a process called inverse design).
An important advantage with the approach is that the network needs to be trained on only a small sampling of the data to approximate the simulation to high precision.
