This module introduces the fundamentals and practical applications of artificial intelligence in a more accessible way. It begins by explaining what AI is in scientific terms and outlines its main branches, including deep learning, natural language processing, and robotics. You will then be guided through the practical workflow of machine learning, with a focus on how data is prepared and used in tools such as Weka and Python. The module also covers key methods for evaluating models, including cross-validation, confusion matrices, precision, and recall, helping you understand how to measure model performance in a meaningful way. Finally, it explores several widely used machine learning algorithms, such as K-nearest neighbors, random forests, and support vector machines, with an emphasis on how they work in practice. Overall, this module aims to build a clear and practical understanding of data-driven decision-making.

This module introduces feature learning principles within a bioinformatics framework, focusing on the prediction of electron transport proteins. It delineates a structured four-stage workflow comprising data collection, feature extraction using metrics like Amino Acid Composition, feature learning via machine learning algorithms, and final model evaluation. Additionally, the curriculum provides practical training in Weka software, demonstrating how to save and load predictive models, perform hyperparameter tuning through controlled experiments in Weka Experimenter, and utilize ArffViewer for converting CSV data into the ARFF format.

This module introduces feature learning principles within a bioinformatics framework, focusing on the prediction of electron transport proteins—key mediators of cellular respiration. It delineates a structured four-stage workflow comprising data collection, feature extraction using metrics such as Amino Acid Composition, Dipeptide Composition, and Position-Specific Scoring Matrix, feature learning via machine learning algorithms including K-Nearest Neighbors and Random Forest, and final model evaluation through cross-validation and independent test sets. Additionally, the curriculum provides practical training in Weka software, demonstrating how to save and load predictive models for reuse without retraining, perform hyperparameter tuning through controlled experiments and paired t-tests in Weka Experimenter, and utilize ArffViewer for converting CSV data into the ARFF format required for classification tasks.

This module transitions from traditional machine learning frameworks to the advanced domain of deep learning and sequence analysis. This week, we will investigate how multi-layered artificial neural networks perform autonomous representation learning, bypassing manual feature engineering to extract intricate biological patterns from large-scale datasets. Our curriculum provides an objective evaluation of Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Natural Language Processing (NLP) workflows, demonstrating their functional application to genomics and proteomics. To ground these scientific concepts, you will complete hands-on technical implementations using the WekaDeeplearning4j software suite and specialized Python packages, mastering network configuration and localized hyperparameter tuning. Please thoroughly review the assigned video lectures and documentation before starting your practical exercises.

This module focuses on scientific communication in bioinformatics, guiding learners through the process of transforming bioinformatics analyses into a publishable research paper. You will explore the standard bioinformatics workflow—from data collection and feature extraction to model development and prediction—and learn how these steps are presented in a research paper. The module also covers dataset construction, feature representation, data visualization, and scientific reporting, using a published bioinformatics paper as a practical example.