Machine Learning for Biologists (with Python)

Take a deep dive into Machine learning - the terminology, statistics, visualization, model fitting, estimation and selection with Python  

What will you learn :

Basic terminology – ML, model, feature, response/target variable

Dataset – Iris and Cancer Dataset – Kaggle database   

Basic statistics behind ML – Linear and Logistic regression

Starting with Google colab

ML paradigms – Supervised and Unsupervised Learning 

Python libraries for ML – Pandas, Numpy, Scikitlearn 

Supervised Learning – Steps in Supervised learning – Data Partition – Training set, Test Set, Cross Validation, Model evaluation, Model selection - Examples of Linear and Logistic Regression

Unsupervised Learning – Scaling, determining optimum clusters, K means clustering, decision tree, random forest

Generics – Feature extraction and selection methods 

Conclusions and comments 

Machine Learning with R

(Recordings available) 

Take a deep dive into Machine learning - the terminology, statistics, visualization, model fitting, estimation and selection with R 

Days 1 - 3 : Introduction to AI and Machine Learning

   Day 1 - Brief introduction to R, RStudio, dplyr

   Day 2 - Fundamentals of Machine Learning

 Day 3 - Exploratory data analysis and visualization with ggplot2

Days 4 - 7 : Machine Learning with Caret and mlbench

Supervised Learning - Linear and Logistic Regression Data Partitioning, Training and test set, model fitting, Cross validation, Model evaluation, Model Selection, Confusion matrix

Unsupervised learning - Scaling, Determining optimal number of clusters, Elbow plot, Within-Cluster Sum of Squares (WSS), K-means-Clustering, model evaluation

NGS Fundamentals and Data Analysis 

DNASeq - Variant  Calling 

(Recordings available)

This workshop offers a comprehensive introduction to the field of Next-Generation Sequencing (NGS) data processing and genomic variant identification. The curriculum begins by exploring modern sequencing technologies - Short Read Sequencing - Illumina and Ion Torrent & Long Read - Nanopore and Sequel, and fundamental bioinformatics tools, such as the Linux command line and the Conda package manager. Participants will learn to navigate essential genomic data formats and perform critical quality control procedures to ensure data integrity. The latter portion of the course emphasizes practical skills in reference based mapping of short reads and the discovery of genetic variants. Finally, the program concludes with methods for genome data visualization, allowing researchers to effectively interpret and display their findings.

Be my guest on this journey most remarkable as i take you through the basics of NGS Data Analysis 

R FOR BIOLOGISTS - Level I

Begin from Scratch

(Recordings available)

Biology today is fast transforming into data-science - curtsey the high-throughput technologies. Manual analysis of this data is neither feasible nor possible anymore. With the data deluge in Life Science Research, programming is fast becoming  a desirable and essential skill, even for wet lab researchers, for data wrangling, analysis,  and data visualization.  

The 'R for Biologists - Level I '  curriculum is specially designed for researchers with no coding background and outlines a structured eight-day (~10 hours) training program. 

The first half of the course focuses on foundational coding skills, covering essential topics such as data types, data structures, control flow, and file management. Students learn to manipulate biological sequences and perform statistical testing before moving on to more complex tools. The latter portion of the syllabus introduces specialized libraries like Dplyr and ggplot2 to facilitate high-level data analysis and visualization. Finally, the course covers bioinformatics-specific packages such as Bioconductor to equip learners with the ability to process Next-Generation Sequencing data.