Current and Future Aspects of AI In Drug Discovery Design and Development.pptx

Current and Future Aspects of AI In Drug Discovery Design and Development

1. Current and Future Aspects of AI In Drug Discovery Design and Development Prof. Dr . Basavaraj Nanjwade M. Pharm., PhD Assist. Vice President and Plant Head KJD Pharmaceuticals Pvt Ltd BIDAR-585402 11/23/2021 1 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

2. CONTENT AI in Drug Discovery AI in Drug Design AI in Drug Development AI in Clinical Trial AI in Pharmaceutical manufacturing AI in QA & QC AI in Pharmaceutical Marketing AI in Disease Diagnosis AI in Disease Prevention AI in Epidemic Prediction AI in Remote Monitoring 11/23/2021 2 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

3. Artificial Intelligence Artificial Intelligence (AI) plays a pivotal role in drug discovery. In particular artificial neural networks such as deep neural networks or recurrent networks drive this area. Numerous applications in property or activity predictions like physicochemical and ADMET properties have recently appeared and underpin the strength of this technology in quantitative structure-property relationships (QSPR) or quantitative structure-activity relationships (QSAR). 11/23/2021 3 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

4. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 4 Artificial Intelligence

5. AI in Drug Discovery Classical pharmacology drug discovery Reverse pharmacology drug discovery Target based drug discovery Polypharmacology Chemical synthesis Drug repurposing Drug screening 11/23/2021 5 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

6. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 6 AI in Drug Discovery

7. Future Aspects of AI Develop Assay Develop Assay Apoptosis Assay Drug Induced Liver Assay Autophagy Assay Endocrine Profiling Assay Cell Cycle Assay In situ Hybridization Assay Cell Health Profiling Assay Invasion Assay Cell spreading Assay Micronucleus Assay Cell Toxicity Assay Micronucles Genotoxicity Assay Cell Viability Assay Mitotic Index Assay Colony Formation Assay Multiple Target Translocation Assay Comet Assay Neurite Detection Assay Comet Genotoxicity Assay Neurotoxicity Profiling Assay Cytoskeletal Rearrangement Assay Organelle Health Assay DNA Damage Assays Proliferation and Cell Death Assay Receptor Internalization Assay Single Target Translocation Assay Synaptogenesis Assay Transfection Efficiency Assay 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 7

8. Classical pharmacology drug discovery Classical pharmacology, also known as forward pharmacology or phenotypic drug discovery (PDD). Chemical libraries of synthetic small molecules. Natural products or extracts to identify substances that have desirable therapeutic effect. https://www.chembridge.com  11/23/2021 8 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

9. Reverse pharmacology drug discovery Reverse pharmacology is the science of integrating documented clinical/experiential hits, into leads by transdisciplinary exploratory studies and further developing these into drug candidates by experimental and clinical research. 11/23/2021 9 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

10. Target based drug discovery The approach is not limited to small-molecule drug discovery. Antibody drugs and other protein biologics are generally discovered through target-based approaches, and The discovery of gene therapy and nucleic acid-based therapeutics is inherently target based as well. 11/23/2021 10 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

11. Classical pharmacology and Target Based Drug Discovery Target-based approach Phenotypic approach Cost Lower Higher Speed Rapid Moderate Ease of SAR Easier More difficult Translatability Depends on other evidence Presumed higher in general Molecular Target Known May not be identified 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 11

12. Polypharmacology Polypharmacology is the design or use of pharmaceutical agents that act on multiple targets or disease pathways. Despite scientific advancements and an increase of global R&D spending, drugs are frequently withdrawn from markets. This is primarily due to their side effects or toxicities. Designing biospecific drug molecules Designing multitarget drug molecule 11/23/2021 12 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

13. Designing biospecific drug molecules Antibodies (Abs) containing two different antigen-binding sites in one molecule are called bispecific . 11/23/2021 13 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

14. Designing multitarget drug molecule The use of multitarget drugs is promising as it lowers the possibility of the disease to evolve into a drug-resistant phenotype. 11/23/2021 14 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

15. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 15 Designing multitarget drug molecule

16. Chemical synthesis Prediction of reaction yield Prediction of retrosynthesis pathways Developing insights into reaction mechanisms Designing synthetic route 11/23/2021 16 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

17. Prediction of reaction yield 11/23/2021 17 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 Chemical syntheses in terms of resource and energy efficiency, product selectivity, operational simplicity, and health and environmental safety.

18. Prediction of retrosynthesis pathways Retrosynthetic pathways is a technique for solving problems in the planning of organic syntheses. 11/23/2021 18 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

19. Developing insights into reaction mechanisms Reaction mechanism, or reaction pathway, describes the successive steps at the molecular level that take place in a chemical reaction. In each step, molecular bonds are either created or broken. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 19

20. Designing synthetic route The simplest synthesis of a molecule is one in which the target molecule can be obtained by using a readily available starting material for a single reaction that converts it to the desired target molecule.  11/23/2021 20 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

21. Drug repurposing Identification of therapeutic target Prediction of new therapeutic use 11/23/2021 21 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

22. Identification of therapeutic target Identifying a biological target that is ‘ druggable ’ – a target is termed ‘ druggable ’ if its activity (behavior or function) can be modulated by a therapeutic – whether it be a small molecule drug, or biologic. Proteins and nucleic acids are both examples of biological targets. 11/23/2021 22 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

23. The target has a confirmed role in the pathophysiology of a disease and/or is disease-modifying. Target expression is not evenly distributed throughout the body. The target’s 3D-structure is available to assess druggability . The target is easily ‘assayable’ enabling high-throughput screening. The target possesses a promising toxicity profile, potential adverse effects can be predicted using phenotypic data. The proposed target has a favorable intellectual property (IP) status. (relevant for pharma companies) 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 23 Identification of therapeutic target

24. Prediction of new therapeutic use Machine-learning methods are particularly suited to predictions based on existing data, but precise predictions about the distant future are often fundamentally impossible.  11/23/2021 24 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

25. Drug screening Prediction of toxicity Prediction of bioactivity Prediction of physicochemical property Identification and classification of target cells 11/23/2021 25 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

26. Prediction of toxicity The prediction of the toxicity of any drug molecule is vital to avoid toxic effects. Cell-based in vitro assays are often used as preliminary studies, followed by animal studies to identify the toxicity of a compound, increasing the expense of drug discovery. Several web-based tools, such as LimTox , pkCSM , admetSAR , and Toxtree , are available to help reduce the cost 11/23/2021 26 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

27. Prediction of bioactivity The efficacy of drug molecules depends on their affinity for the target protein or receptor. Drug molecules that do not show any interaction or affinity towards the targeted protein will not be able to deliver the therapeutic response. 11/23/2021 27 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

28. Prediction of physicochemical property Physicochemical properties, such as solubility, partition coefficient ( logP ), degree of ionization, and intrinsic permeability of the drug, indirectly affect its pharmacokinetics properties and its target receptor family and, hence, must be considered when designing a new drug. Different AI-based tools can be used to predict physicochemical properties. 11/23/2021 28 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

29. Identification and classification of target cells Target identification begins with identifying the function of a possible therapeutic target (gene/protein) and its role in the disease. Target cells are erythryoctes with an increased cell membrane-to-volume ratio, due either to gain of membrane lipids or to a reduction in cell volume. 11/23/2021 29 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

30. AI tools for Drug Discovery DeepChem : https://github.com/deepchem/deepchem DeepTox : www.bioinf.jku.at/research/DeepTox DeepNeuralNetQSAR : https://github.com/Merck/DeepNeuralNet-QSAR ORGANIC: https://github.com/aspuru-guzik-group/ORGANIC PotentialNet : https://pubs.acs.org/doi/full/10.1021/acscentsci.8b00507 Hit Dexter: http://hitdexter2.zbh.uni-hamburg.de DeltaVina : https://github.com/chengwang88/deltavina Neural graph fingerprint: https://github.com/HIPS/neural-fingerprint AlphaFold : https://deepmind.com/blog/alphafold Chemputer : https://zenodo.org/record/1481731 11/23/2021 30 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

31. AI in Drug Design Computer aided drug design Structure based drug design Predicting 3D structure of target protein Predicting drug-protein interactions Determining drug activity De novo drug design 11/23/2021 31 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

32. Computer aided drug design (CADD) CADD helps scientists in minimizing the synthetic and biological testing efforts by focussing only on the most promising compounds. Identification and optimization of new drugs using leverage of chemical and biological information about targets and/or ligands . 11/23/2021 32 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

33. Structure based drug design Once a target has been identified, it is necessary to obtain accurate structural information. There are three primary methods for structure determination that are useful for drug design:  X-ray crystallography, NMR, and homology modeling . The evaluation of structures from each method will be discussed. 11/23/2021 33 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

34. Predicting 3D structure of target protein Currently, the main techniques used to determine protein 3D structure are  X-ray crystallography and nuclear magnetic resonance (NMR) . In X-ray crystallography the protein is crystallized and then using X-ray diffraction the structure of protein is determined. 11/23/2021 34 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

35. Predicting drug-protein interactions Drug - Protein Interaction, In order for a  drug  to act upon a receptor or be metabolized by an enzyme it must first be bound to receptor  protein  or enzyme. 11/23/2021 35 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

36. Determining drug activity Physiologic differences between men and women affect drug activity, including pharmacokinetics and pharmacodynamics . Pharmacokinetics in women is affected by lower body weight, slower gastrointestinal motility, less intestinal enzymatic activity, and slower glomerular filtration rate. Because of delayed gastric emptying, women may need to extend the interval between eating and taking medications that must be abs orbed on an empty stomach. 11/23/2021 36 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

37. De novo drug design In de novo design, to identify amino acid sequences folding into proteins with desired functions is the ultimate objective. There are various approaches predicting protein structure including comparative modeling and fold recognition. In proteins, sequence similarity implies structural similarity. 11/23/2021 37 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

38. AI in Drug Development QbD GMP cGMP Stability issues Dissolution Porosity 11/23/2021 38 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

39. AI can assist in selecting only a specific diseased population for recruitment in Phase II and III of clinical trials by using patient-specific genome– exposome profile analysis,which can help in early prediction of the available drug targets in the patients selected. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 39 AI in Clinical Trial

40. AI in Clinical Trial Phase I Phase II Phase III Phase IV Subject enrolments/selection Patient dropout Monitoring of trial 11/23/2021 40 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

41. Subject enrolments/selection The enrolment of patients takes one-third of the clinical trial timeline. The success of a clinical trial can be ensured by the recruitment of suitable patients, which otherwise leads to 86% of failure cases. 11/23/2021 41 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

42. Patient dropout Drop out of patients from clinical trials accounts for the failure of 30%of the clinical trials, creating additional recruiting requirements for the completion of the trial, leading to a wastage of time and money. 11/23/2021 42 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

43. Monitoring of trial This can be avoided by close monitoring of the patients and helping them follow the desired protocol of the clinical trial 11/23/2021 43 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

44. AI in pharmaceutical manufacturing Automated manufacturing Personalized manufacturing Correlating manufacturing errors to set parameters Quality control Predictive maintenance Waste reduction Design optimization Process automation 11/23/2021 44 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

45. Automated manufacturing The pharmaceutical industry is increasingly making use of robotics to automate specific processes in drug development, including drug screening, anti-counterfeiting and manufacturing tasks.  This allows for the study of specific interactions between a particular drug and its protein target at the atomic level. 11/23/2021 45 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

46. Personalized manufacturing Custom manufacturing is the process of designing, engineering, and producing goods based on a customer's unique specifications, including build to order (BTO) parts, one-offs, short production runs, as well mass customization. 11/23/2021 46 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

47. Correlating manufacturing errors to set parameters Additive Manufacturing (AM) is a process that is based on manufacturing parts layer by layer in order to avoid any geometric limitation in terms of creating the desired design. In the early stages of AM development, the goal was just creating some prototypes to decrease the time of manufacturing assessment.  11/23/2021 47 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

48. Quality control The lack of oversight and proper quality control imposes severe repercussions for any drug manufacturer and their customers. The chance of contamination puts customers at risk for potentially life-threatening allergic reactions. The impact of those errors can result in litigation, reputational harm, and FDA regulatory action. 11/23/2021 48 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

49. Predictive maintenance Predictive maintenance refers to the use of data-driven, proactive maintenance methods that are designed to analyze the condition of equipment and help predict when maintenance should be performed. 11/23/2021 49 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

50. Waste reduction Waste reduction (or prevention) is the preferred approach to waste management because waste that never gets created doesn't have waste management costs. An example of waste reduction is reducing unnecessary packaging from manufactured products and produce. 11/23/2021 50 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

51. Design optimization Design optimization is an engineering design methodology using a mathematical formulation of a design problem to support selection of the optimal design among many alternatives. 1. Variables: Describe the design alternatives 2. Objective: Elected functional combination of variables (to be maximized or minimized) 3. Constraints: Combination of Variables expressed as equalities or inequalities that must be satisfied for any acceptable design alternative 4. Feasibility: Values for set of variables that satisfies all constraints and minimizes/maximizes Objective. 11/23/2021 51 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

52. Process automation Robotic process automation (RPA) software automatically handles manual, repetitive, time-consuming, and highly structured tasks such as data entry and back-office functions. 11/23/2021 52 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

53. Quality Assurance and Control Understand critical process parameters Guide future production cycle Regulation of in-line quality Ensure QA with aid of ELN and other techniques In order to simplify the effort of visual error detection in the productive pharmaceutical industry and to reduce error rates, elunic developed a system for automatic quality assurance with the help of deep learning algorithms. 11/23/2021 53 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

54. Understand critical process parameters A Critical Process Parameter (CPP) is a term used in pharmaceutical production for process variables which have an impact on a critical quality attribute (CQA) and, therefore, should be monitored or controlled to ensure the drug product obtains the desired quality. 11/23/2021 54 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

55. Guide future production cycle 11/23/2021 55 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

56. Regulation of in-line quality The in-line quality control process incorporates inspection points across the production line. These points inspect the product for quality in terms of various standards or specifications.  Process control Control charts Product quality control, Process control 11/23/2021 56 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

57. Ensure QA with aid of ELN and other techniques Electronic Lab Notebooks (ELN), Laboratory Information Management Systems (LIMS), Chromatography Data Systems (CDS), and Laboratory Execution Systems (LES). 11/23/2021 57 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

58. AI in Pharmaceutical Marketing Market positioning Market prediction and analysis Product costing 11/23/2021 58 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

59. Market positioning Positioning refers to the place that a brand occupies in the minds of the customers and how it is distinguished from the products of the competitors and different from the concept of brand awareness. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 59

60. Market prediction and analysis Technical analysis focuses on analyzing historical stock prices to predict future stock values (i.e. it focuses on the direction of prices). A market analysis is a quantitative and qualitative assessment of a market. It looks into the size of the market both in volume and in value, the various customer segments and buying patterns, the competition, and the economic environment in terms of barriers to entry and regulation. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 60

61. Product costing Product cost refers to the costs incurred to create a product. These costs include direct labor, direct materials, consumable production supplies, and factory overhead. Product cost can also be considered the cost of the labor required to deliver a service to a customer. 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 61

62. AI in Disease Diagnosis Doctors can use advanced Machine Learning systems to collect, process, and analyze vast volumes of patients’ healthcare data. Healthcare providers around the world are using ML technology to store sensitive patient data securely in the cloud or a centralized storage system. This is known as electronic medical records (EMRs). ML systems can use the data stored in EMRs to make real-time predictions for diagnosis purposes and suggest proper treatment to patients.  11/23/2021 62 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

63. AI in Disease Prevention Pharma companies can use AI to develop cures for both known diseases like Alzheimer’s and Parkinson’s and rare diseases. Generally, pharmaceutical companies do not spend their time and resources on finding treatments for rare diseases since the ROI is very low compared to the time and cost it takes to develop drugs for treating rare diseases. 11/23/2021 63 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

64. AI in Epidemic Prediction AI and ML are already used by many pharma companies and healthcare providers to monitor and forecast epidemic outbreaks across the globe. Such AI/ML models become especially useful for underdeveloped economies that lack the medical infrastructure and financial framework to deal with an epidemic outbreak. 11/23/2021 64 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

65. AI in Remote Monitoring Telemedicine consults, remote monitoring devices, and AI are all evolving to provide healthcare teams new ways to gain deeper levels of patient information that enhance diagnosis and treatment decisions, while also fostering client relationships. 11/23/2021 65 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090

66. Pharmaceuticals with AI Organizations 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 66 Drug Discovery Today Volume 26, Number 1 January 2021

67. Artificial Intelligence Software AI Tools Functionality Google Cloud Machine Learning Azure Machine Learning TensorFlow Machine Learning H2O AI Machine Learning Cortana Virtual Assistant IBM Watson Question-answering system. Infosys Nia Machine Learning Chatbot . Amazon Alexa Virtual Assistant Google Assistant Virtual Assistant 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 67

68. THANK YOU E-mail: [email protected] Mob#: [phone] 11/23/2021 Acharya & BM Reddy College of Pharmacy, Bengaluru-560090 68