Machine learning and big data problems in bioinformatics

Title: 

Machine learning and big data problems in bioinformatics

Abstract:

This two-hour tutorial will cover some interesting machine learning problems in bio molecular networks, describe some data mining and machine learning problems in genomic sequence compression, and discuss some challenging machine learning problems in the genome editing area.

Overview describing the topic

• Disease and gene association networks, protein-protein interaction graphs, small RNAs and messenger RNAs regulation networks; missing edge prediction, elimination of false positive edges; ensemble learning algorithms for these prediction.
• Unsupervised clustering of long sequences; long pattern mining from long sequences; lossless compression based on the long patterns.
• Genome editing on-target predictions and genome editing off-target predictions by machine learning algorithms.

Research areas and prior knowledge required of potential audience

• Basic machine learning, artificial intelligence, or data mining knowledge
• Basic biology background.

 A brief outline of the tutorial structure showing the tutorial’s core content

• This tutorial consists of three parts (two hours in total)
• The first part introduces biomolecular networks and commonly used machine learning and data mining algorithms for the prediction of false positives or missing edges or both.
• The second part introduces unsupervised learning algorithms which have been used for genome sequence compression.
• The third part presents feature space construction methods and boosting algorithms for the prediction of sgRNAs in the genome editing systems.

Short CV of the speaker:

Dr. Jinyan Li is a Professor of Data Science and Program Leader of Bioinformatics at the Advanced Analytics Institute, University of Technology Sydney, Australia. He has been actively working on data mining and bioinformatics for 20 years. He has published 220 papers, including 120 papers in the prestigious journals of data mining, machine learning, and computational biology. He is widely known for his pioneering research on the theories and algorithms of emerging patterns. One of these papers has received 1200 Google Scholar citations. Jinyan has a Bachelor degree of Science (Applied Mathematics) from National University of Defense Technology (China), a Master degree of Engineering (Computer Engineering) from Hebei University of Technology (China), and a PhD degree (Computer Science) from the University of Melbourne (Australia). More details of his research can be found athttp://www.uts.edu.au/staff/jinyan.li