Artificial Intelligence in Chemistry Muhammadd Zubair BSCHEM 3rd
Contents Introduction Applications in various fields Applications in chemistry Limitations 2
Introduction John MC Carthy_______1956 Branch of computer science Deals with Machine Learning Process Capable of performing different tasks that typically require human intelligence. Replication of human intelligence in machine 3
Applications In Various Fields
Applications in Technology Industry Artificial Intelligence helps in: Collecting & analyzing data at a low cost. Having a safe working environment 5 Applications in Engineering Field Artificial Intelligence helps in: Natural language processing Reasoning problems Making strategy
Applications in Chemical Field Artificial Intelligence helps in: Retrosynthesis Prediction of reaction outcomes Molecular designing Waste-water treatment Nano-technology Drug discovery Molecular property prediction 6
Retrosynthesis It is a technique in which a target molecule is transformed into a simple precursor. This process is continue until the starting molecule is obtained. AI tool use for basically complex molecules e.g. Hurculean B ___32-stereo-centres______ 47 reaction required. 7
Contd... 8
Prediction of Reaction Outcomes The outcomes of the reaction can be predicted by using two steps: By applying a forward reaction template to the reactant molecule. By finding the reaction products that is formed in major amount by using machine learning. For example: A+B C (50%) A+B AB D (30%) 9
Molecular Designing Machine learning based tools are helpful in the discovery of : Molecular pattern Properties e.g. M.P, B.P, Solubility, & Stability etc. Size of molecule PRESENTATION TITLE 10
Waste-water Treatment Eutrophication is important issue. Water Pollution increase day by day. AI helpful by taking: Data from biological stage & provides behaviour of bioreactor. Provides operations that is useful in water treatment. Also use some biological, physical & chemical properties. 11
Nano-technology AI plays a significant role in nano-technology Discovery of novel materials with desired properties at nanoscale. Optimize & control nanofabrication processes. Can control & program nanorobots for various applications. Analyze the massive amount of data generated in nano-technology experiments 12
Drug Discovery It is difficult to make drug from 10-100000 molecules. To solve this problem we Use AI. Targeted Drug delivery Predicting Drug-Drug Interaction. Drug release control. Quality control. 13
Molecular property Prediction AI helps in prediction of: Chemoinformatics Protein folding Property optimization Drug repurposing 14
Limitations There are some limitations : Data quality Generalization Computational resources Ethical concerns 15 Lack of domain knowledge Overfitting Expensive implementation Safety risks