major phase 3 report of rvce of interaction based ligand

Interaction based ligand design

1. Interaction-Based Ligand Design for Targeted Drug Discovery ASHWAT KUMAR 1RV20BT005 DEEPAK KUMAR 1RV20BT011 MAJOR PROJECT PHASE 3 18BT74 Department Of Biotechnology 2023-24 DEPARTMENT OF BIOTECHNOLOGY I-15

2. CONTENTS Workflow Methodology Results Expected Outcome Work Pending References Department Of Biotechnology 2023-24 I-15

3. Workflow Department Of Biotechnology 2023-24 Steps: - Data Collection: Gather data from ChEMBL database. - Data Preprocessing: Clean, normalize, and handle missing values. - Descriptor Calculation: Calculate Lipinski descriptors and PaDEL descriptors. - Exploratory Data Analysis (EDA): Perform distribution analysis, identify outliers, and visualize data. - Bioactivity Classification: Classify compounds based on IC50 values and convert to pIC50. - Model Building: Train Random Forest Regressor and make predictions. - Model Evaluation: Evaluate model using MSE and R², and visualize results. I-15

4. Methodology Department Of Biotechnology 2023-24 Data Collection Source of Data: - ChEMBL database: A manually curated database of bioactive molecules with drug-like properties. Types of Data: - Ligand structures (SMILES, SDF files) - Target protein structures (PDB files, sequence data) - Bioactivity data (IC50 values, pIC50 values) Data Example: - Example entry for Acetylcholinesterase: - Organism: Homo sapiens - Target ChEMBL ID: CHEMBL220 - Target Type: SINGLE PROTEIN - Tax ID: 9606 I-15

5. Methodology Department Of Biotechnology 2023-24 Data Preprocessing Steps: - Data Cleaning: Removing duplicates and irrelevant entries. - Normalization: Scaling the data to ensure consistency. - Handling Missing Values: Using imputation techniques to fill missing data. Tools Used: - Python, Pandas Steps Taken: - Data Cleaning: Ensured no duplicates or irrelevant entries remained. - Normalization: Applied to maintain consistent scales across the dataset. - Handling Missing Values: Imputation methods used to address gaps in the data. I-15

6. Methodology Department Of Biotechnology 2023-24 Descriptor Calculation Lipinski's Rule of Five: - Key descriptors: Molecular Weight (MW), LogP , NumHDonors , NumHAcceptors . - Importance: Helps in identifying drug-like properties of compounds. PaDEL -Descriptor: - Calculation of molecular descriptors and fingerprints. Steps Taken: - Calculated key descriptors using RDKit and PaDEL -Descriptor. - Ensured all descriptors were accurately recorded for each compound. I-15

7. Methodology Department Of Biotechnology 2023-24 Exploratory Data Analysis (EDA) Steps: - Distribution Analysis: Examining the spread and central tendency of data. - Identification of Outliers: Detecting anomalies and data quality issues. - Visualization: Creating histograms, scatter plots, and KDE (Kernel Density Estimate) plots. Steps Taken: - Performed thorough distribution analysis to understand data spread. - Identified and handled outliers to maintain data integrity. - Created comprehensive visualizations for better data insights. I-15

8. Methodology Department Of Biotechnology 2023-24 Bioactivity Classification Classification Criteria: - Active: IC50 < 1000 nM - Inactive: IC50 > 10,000 nM - Intermediate: IC50 between 1000 nM and 10,000 nM IC50 and pIC50: - Measure the concentration of a compound where 50% of its maximal inhibitory effect is observed. - Convert IC50 values to pIC50 for a more convenient scale. Steps Taken: - Classified compounds into active, inactive, and intermediate categories. - Converted IC50 values to pIC50 to streamline data analysis. I-15

9. Results Department Of Biotechnology 2023-24 Data Visualization Scatter Plot of MW vs LogP : - Visualizing the relationship between molecular weight (MW) and LogP values. - Show how active and inactive compounds are distributed in the chemical space. Box Plots: - Comparing pIC50 values between active and inactive compounds. - Highlighting differences in bioactivity classes. Visualizations Created: - Scatter Plot: Showed correlation between MW and LogP . - Box Plots: Provided clear comparison of bioactivity classes. I-15

10. Results Department Of Biotechnology 2023-24 I-15

11. Results Department Of Biotechnology 2023-24 Statistical Analysis Mann-Whitney U Test: - Comparing the distribution of pIC50 values between active and inactive compounds. - Non-parametric test to determine if there are significant differences. Findings: - Significant differences were found between the distributions of active and inactive compounds. - Statistical evidence supported the bioactivity classification. Steps Taken: - Performed Mann-Whitney U Test on pIC50 values. - Analyzed test results to confirm statistical significance. I-15

12. Results Department Of Biotechnology 2023-24 I-15

13. Results Department Of Biotechnology 2023-24 Model Building Algorithm: Random Forest Regressor - Suitable for handling large datasets and capturing complex relationships. - Robust against overfitting and easy to interpret. Training and Testing: - Splitting the data into training and testing sets (80/20 ratio). - Training the model on the training set and making predictions on the testing set. Steps Taken: - Chose Random Forest due to its robustness and performance. - Trained and tested the model to predict pIC50 values. I-15

14. Results Department Of Biotechnology 2023-24 I-15

15. Results Department Of Biotechnology 2023-24 Model Evaluation Metrics: - Mean Squared Error (MSE): Measure of the average squared difference between actual and predicted values. - R-Squared (R²): Proportion of the variance in the dependent variable that is predictable from the independent variables. Visualization: - Scatter Plot of Experimental vs Predicted pIC50 Values. - Show the model's performance visually. Findings: - Model achieved high R-Squared value, indicating good predictive power. - Scatter plot showed strong correlation between experimental and predicted values. I-15

16. Results Department Of Biotechnology 2023-24 I-15

17. Expected Outcomes Department Of Biotechnology 2023-24 Improved Prediction Accuracy: The model is expected to achieve higher accuracy in predicting optimal drug-target interactions compared to traditional methods. Accelerated Drug Development: The model's data-driven approach is anticipated to streamline drug development timelines, facilitating faster identification of promising drug candidates. Deeper Insights into Molecular Mechanisms: The model will provide deeper insights into molecular interactions, enabling a comprehensive understanding of disease pathways and therapeutic targets I-15

18. Work Pending Department Of Biotechnology 2023-24 Front-End Application: Development of a user-friendly front-end application for researchers to interact with the model. Novel Ligand Generation: Generating novel ligand compounds based on model predictions and validating their efficacy through experimental collaborations. Further Optimization: Continuous refinement of the model to enhance its predictive power and generalization capabilities. I-15

19. References Department Of Biotechnology 2023-24 H. Chen, O. Engkvist , Y. Wang, M. Olivecrona , and T. Blaschke , "The rise of deep learning in drug discovery," Drug Discovery Today, vol. 23, no. 6, pp. 1241-1250, 2018. G. Schneider, "Automating drug discovery," Nature Reviews Drug Discovery, vol. 17, pp. 97-113, 2018. E. Herrero and F. J. Luque , "Merging Ligand-Based and Structure-Based Methods in Drug Discovery: An Overview of Combined Virtual Screening Approaches," Molecules, vol. 25, no. 20, p. 4723, Oct. 2020, doi : 10.3390/molecules25204723. D. M. Hawkins, "The Problem of Overfitting," Journal of Chemical Information and Modeling, vol. 44, no. 1, pp. 1-12, 2004, doi : 10.1021/ci0342472. J. Ma, R. Sheridan, and A. Liaw , "Building Predictive Models for Drug Discovery Using the Supervised Learning Approach," Journal of Chemical Information and Modeling, vol. 55, no. 1, pp. 21-28, 2015, doi : 10.1021/ci500595g. G. G. Glen and A. I. Tropsha , "Novel quantitative structure-activity relationship (QSAR) methods: The reality and the myth," Journal of Chemical Information and Modeling, vol. 42, no. 6, pp. 1474-1480, 2002, doi : 10.1021/ci020245e. I-15

20. References Department Of Biotechnology 2023-24 M. W. Young, "Computational Approaches in Drug Discovery," Journal of Medicinal Chemistry, vol. 60, no. 6, pp. 2356-2366, 2017, doi : 10.1021/acs.jmedchem.7b00252. A. L. Hopkins, "Network pharmacology: the next paradigm in drug discovery," Nature Chemical Biology, vol. 4, no. 11, pp. 682-690, 2008, doi : 10.1038/nchembio.118. J. P. Hughes, S. Rees, S. B. Kalindjian , and K. L. Philpott, "Principles of early drug discovery," British Journal of Pharmacology, vol. 162, no. 6, pp. 1239-1249, 2011, doi : 10.1111/j.1476-5381.2010.01127.x. H. Chen, O. Engkvist , Y. Wang, M. Olivecrona , and T. Blaschke , "The rise of deep learning in drug discovery," Drug Discovery Today, vol. 23, no. 6, pp. 1241-1250, 2018. G. Schneider, "Automating drug discovery," Nature Reviews Drug Discovery, vol. 17, pp. 97-113, 2018. E. Herrero and F. J. Luque , "Merging Ligand-Based and Structure-Based Methods in Drug Discovery: An Overview of Combined Virtual Screening Approaches," Molecules, vol. 25, no. 20, p. 4723, Oct. 2020, doi : 10.3390/molecules25204723. I-15

21. References Department Of Biotechnology 2023-24 G. Schneider and U. Fechner, "Computer-based de novo design of drug-like molecules," Nat. Rev. Drug Discov ., vol. 4, pp. 649–663, 2005. doi : 10.1038/nrd1799. M. Popova, O. Isayev, and A. Tropsha , "Deep reinforcement learning for de novo drug design," Sci. Adv., vol. 4, no. 7, pp. eaap7885, 2018. doi : 10.1126/sciadv.aap7885. P. Schneider et al., "Rethinking drug design in the artificial intelligence era," Nat. Rev. Drug Discov ., vol. 19, pp. 353–364, 2019. doi : 10.1038/s41573-019-0012-2. G. M. Maggiora , "On outliers and activity cliffs — why QSAR often disappoints," J. Chem. Inf. Model., vol. 46, p. 1535, 2006. doi : 10.1021/ci060138h. N. Bosc et al., "Large scale comparison of QSAR and conformal prediction methods and their applications in drug discovery," J. Cheminform ., vol. 11, p. 4, 2019. doi : 10.1186/s13321-019-0330-1. I-15

22. THANK YOU Department Of Biotechnology 2023-24 I-15