Artificial Intelligence (AI) and its role in Drug Target Identification

Drug Target Identification and AI

Target as the name suggests is the receptor where the drug binds to, to act on the disease or condition. The target might be an established cellular or molecular structure involved in the pathology of interest or might be a new target in the study.

Target plays a vital role in the process of drug development. 

Why is finding a drug target essential in the drug development process?

The targets are mainly groups of proteins. These proteins are made up of amino acids, which in turn are organic compounds that contain amino and carboxyl functional groups. These targets form bonding with the chemical structure of the drug leading to further biological reactions in the biochemical pathway.

This makes the study of the structure of proteins of the target a vital step in identifying the target. In traditional methods, the target is identified using various chemical reactions between the stipulated target and the desired drug molecule. The target which binds with the drug to show the desired action in the disease progression was termed as the required target.

The target can be identified using target deconvolution or target discovery method.

The target deconvolution method involves exposing the desired cells or the tissues to the study molecule to understand how the molecule reacts to achieve any desired outcome. These targets are established targets, whose data is available in various drug banks. Using the predetermined targets, the drugs are tested using various techniques. 

Few of the techniques that involve the deconvolution method are:

  • Affinity chromatography
  • Protein microarray
  • Expression – cloning
  • Biochemical suppression
  • Reverse Transfected cell microarray

The target discovery method does not involve the pre-established targets. New targets have been discovered for the study of the drug molecules. This is an important step. As there are newer inventions of drugs and there is a change in the behavior patterns of the pathogens, there is a need to discover newer targets rather than depending on the old targets.

Some of the common techniques used are:

  • DNA microarray
  • siRNA
  • System biology
  • Study of Existing Drugs

Drawbacks of traditional methods of Target Identification:

  1. The traditional method mainly uses trial and error methods. 
  2. Even though there is advancement in the method of finding the target, the chances of the target being rejected are very high. 
  3. The cost and time involved in this process is huge. Normally, the drug discovery takes around 11 to 15 years with about cost involvement of around 2.6$ billion [1]
  4. Many small-scale manufacturing companies, though having potential resources cannot afford to develop new drugs due to the involvement of huge costs.
  5. It is really the need of the hour to develop faster solutions when pandemic-like situations arise. 

These drawbacks can be easily overcome by the combination of traditional methods and artificial intelligence. Using artificial intelligence, it becomes easier to understand the 3D-protein structure of the molecules. Hence the best match between the drug molecule and the huge number of proteins becomes very easy. The cost and time involved in testing different established and non-established targets to find the required target can be avoided. Instead testing the sure-shot targets obtained by artificial intelligence paves way for providing precise treatment at a lesser cost and time. 

Artificial intelligence also helps in finding a better match through already established drug banks using a more accurate pattern.

AI drug design algorithm Reinforcement Learning can be successfully deployed for a ‘Simplified molecular-input line-entry system (SMILES) [2]. It is a specification for describing the chemical structure using short ASCII strings or the language that is understood by the machines. 

Few developments in the path of target identification using artificial intelligence are:

The biological reactions in plants and animals take place using different biochemical pathways. The pathway or mechanism of action analysis will help to understand the step or the target in which the drug must react in order to provide the desired condition. The analysis of the pathway using machine learning will help to analyze the reactions in more depth.

The biochemical pathways are the enzyme-mediated reactions where the product of one reaction forms the substrate for another reaction. The enzymes are made up of proteins.

The proteins are made up of amino acids that are expressed by the DNA through the transcription and translation process. In the transcription process, the information from the DNA is transferred to the mRNA which then translates into the particular protein. This entire process of transferring the information stored in the DNA into protein is known as gene expression. miRNA helps in the RNA silencing i.e, negative regulation of gene expression and post-transcriptional regulation of the gene expression. Hence miRNA plays a vital role in determining the biological pathway of the disease condition. 

miRNA target identification in various disease conditions and the various biological processes performed using machine learning will help in finding a precise cure for the disease at the genomic level.

As per the press report, Benvolent and AstraZeneca have partnered for incorporating the use of artificial intelligence in target identification. This would help to identify new drugs for chronic Kidney Disease and Idiopathic Pulmonary Fibrosis. As per the partnership, Benevolent will be using its data integration platform to extend its proprietary Knowledge Graph with multiple AstraZeneca data sets across. [3]

InveniAI® LLC and Kyowa Kirin Co., Ltd have also collaborated. As per the collaboration agreement, Kyowa Kirin Co., Ltd is going to harness the power of InveniAI’s AI-platform, AlphaMeld®, for novel target discovery [4].

With such collaborations and ongoing drug development processes along with artificial intelligence, more effective drugs are sure to enter the market. Since, the target obtained due to artificial intelligence can be made more precise, by fine-tuning, the side effects of the drugs due to action on a wide number of targets can also be reduced.

As the involved cost in drug development will be reduced, there will be a reduction in the market drugs.

I have experience in the healthcare industry for more than ten years. I am a passionate blogger who is interested in writing on healthcare, technology advancements.
follow me
×
I have experience in the healthcare industry for more than ten years. I am a passionate blogger who is interested in writing on healthcare, technology advancements.
follow me
Latest Posts
  • Drug Target Identification and AI
  • Healthcare Advancements Through AI-driven Smartwatches
  • Role of Artificial Intelligence (AI) in Medical Device Development
Share This Post
Written by Prasidha Ramnathan
I have experience in the healthcare industry for more than ten years. I am a passionate blogger who is interested in writing on healthcare, technology advancements.
Have your say!
10

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Thanks for submitting your comment!