Publications by authors named "Muhammad Adnan Pramudito"
This study introduces an innovative computational approach using hybrid machine learning models to predict toxicity across eight critical end points: cardiac toxicity, inhalation toxicity, dermal toxicity, oral toxicity, skin irritation, skin sensitization, eye irritation, and respiratory irritation. Leveraging advanced cheminformatics tools, we extracted relevant features from curated data sets, incorporating a range of descriptors such as Morgan circular fingerprints, MACCS keys, Mordred calculation descriptors, and physicochemical properties. The consensus model was developed by selecting the best-performing classifier-Random Forest (RF), eXtreme Gradient Boosting (XGBoost), or Support Vector Machines (SVM)-for each descriptor, optimizing predictive accuracy and robustness across the end points.
View Article and Find Full Text PDFArticle Synopsis
- - The CiPA initiative is working to better predict the risk of torsades de pointes (TdP) related to drug use by using computational models and explainable artificial intelligence (XAI) to analyze in-silico biomarkers and their impact on drug toxicity.
- - The study created a dataset of 28 drugs and computed twelve in-silico biomarkers, training several machine learning models (like ANN and SVM) to predict toxicity risks, finding that the ANN model performed the best overall in classification accuracy.
- - Using the SHAP method, the researchers were able to identify which biomarkers significantly contributed to the toxicity predictions, but they noted that biomarker selection didn't always enhance model performance, emphasizing the need to evaluate multiple classifiers
A stacking ensemble machine learning model for evaluating cardiac toxicity of drugs based on in silico biomarkers.
CPT Pharmacometrics Syst Pharmacol
December 2024
Article Synopsis
- - This study focuses on improving the assessment of drug-induced torsades de pointes (TdP) risk, which is crucial for drug development due to potential arrhythmias and sudden cardiac death.
- - It introduces a stacking ensemble machine learning model that combines various biomarkers and hERG dynamics, achieving high accuracy in predicting risk levels associated with TdP.
- - The research also looks into the variability among individuals by using data from different human ventricular cell models and identifies critical ion channels that significantly impact TdP risk prediction.
Predicting ventricular arrhythmia Torsade de Pointes (TdP) caused by drug-induced cardiotoxicity is essential in drug development. Several studies used single biomarkers such as qNet and Repolarization Abnormality (RA) in a single cardiac cell model to evaluate TdP risk. However, a single biomarker may not encompass the full range of factors contributing to TdP risk, leading to divergent TdP risk prediction outcomes, mainly when evaluated using unseen data.
View Article and Find Full Text PDFHeart-sound auscultation is one of the most widely used approaches for detecting cardiovascular disorders. Diagnosing abnormalities of heart sound using a stethoscope depends on the physician's skill and judgment. Several studies have shown promising results in automatically detecting cardiovascular disorders based on heart-sound signals.
View Article and Find Full Text PDF