for Bioinformatics, Part I.pdf

1Lecture at Yang Ming National University, Taipei, June 2006 Knowledge Discovery Techniques for Bioinformatics Limsoon Wong Lecture at Yang Ming National University, Taipei, June 2006 Copyright 2006 ? Limsoon Wong 2 Course Plan ? Knowledge discovery essence (1.5 hrs) – Feature generation, feature selection, feature integration – Understanding accuracy – Cross validation ? Machine learning methods (1.5 hrs) – Decision trees – Ensemble classifiers – K-NN – Na?ve Bayes – HMM ? Gene feature recognition (1.5 hrs) – Translation initiation site – Transcription start site ? Gene expr analysis (1.5 hrs) – Gene expr profile classification – Gene expr profile clustering – Extreme sample selection ? Protein subcellular localization recognition (1.0 hrs) – Sorting signals – Amino acid domain compositions ? Data Cleansing (0.5 hrs) – Association rules mining – Errors in databases – Data de-duplication ? Sequence homology interpretation (1.5 hrs) – Guilt by various types of assoc 2Lecture at Yang Ming National University, Taipei, June 2006 For written notes on this lecture, please read chapter 3 of The Practical Bioinformatician, Knowledge Discovery Techniques for Bioinformatics, Part I: Essence of Knowledge Discovery Limsoon Wong Lecture at Yang Ming National University, Taipei, June 2006 Copyright 2006 ? Limsoon Wong 4 Jonathan’s rules : Blue or Circle Jessica’s rules : All the rest What is Knowledge Discovery? Whose block is this? Jonathan’s blocks Jessica’s blocks 3Lecture at Yang Ming National University, Taipei, June 2006 Copyright 2006 ? Limsoon Wong 5 What is Knowledge Discovery? Question: Can you explain how? Lecture at Yang Ming National University, Taipei, June 2006 Copyright 2006 ? Limsoon Wong 6 The Steps of Knowledge Discovery ? Training data gathering ? Feature generation – k-grams, colour, texture, domain know-how, ... ? Feature selection – Entropy, χ2, CFS, t-test, domain know-how... ? Feature integration – SVM, ANN, PCL, CART, C4.5, kNN, ... Some classi