- •LECTURE 9
- •MACHINE LEARNING OVERVIEW
- •LEARNING
- •MAINTAINING A BALANCE
- •A PARTIAL CHARACTERISATION OF LEARNING TASKS
- •MAINTAINING A BALANCE
- •MAINTAINING A BALANCE
- •INDUCTIVE LOGIC PROGRAMMING
- •EXAMPLE LEARNED LP
- •STOCHASTIC LOGIC PROGRAMS
- •AUTOMATED THEORY FORMATION
- •OTHER MACHINE LEARNING METHODS
- •BIOINFORMATICS OVERVIEW
- •FROM SEQUENCE TO STRUCTURE
- •PROBLEM NUMBER ONE
- •PROBLEM NUMBER TWO
- •OTHER AIMS OF BIOINFORMATICS
- •SOME CURRENT
- •A SUBSTRUCTURE SERVER
- •THE SUBSTRUCTURE SERVER
- •USING MEDICAL ONTOLOGIES
- •GENE ONTOLOGY DISCOVERY
- •STUDYING BIOCHEMICAL NETWORKS
- •CLOSED LOOP MACHINE LEARNING
- •FUTURE DIRECTIONS FOR MACHINE LEARNING IN BIOINFORMATICS
- •BIOCHEMICAL PATHWAYS
THE SUBSTRUCTURE SERVER
USING MEDICAL ONTOLOGIES
Use Ontology and ML for database integration
Muggleton and Tamaddoni-Nezhad
Bridge between two disparate databases
LIGAND (biochemical reactions)
Enzyme classification system (EC) = ontology
Automated ontology maintenance
Colton and Traganidas (MSc. Last year)
Gene Ontology (big project)
Use data to find links between GO terms
Equivalence and implication finding using HR
GENE ONTOLOGY DISCOVERY
55%
STUDYING BIOCHEMICAL NETWORKS
Use SLPs to find mappings between genomes
Map function of pairs of homologous proteins
E.g., mouse and human
Homology is probabilistic
Developed SLP learning algorithms
Initial results applying them in biological networks
Work by
Muggleton, Angeloupolos and Watanabe
CLOSED LOOP MACHINE LEARNING
Active learning
Information theoretic algorithm designs and chooses the most informative and lowest cost experiments to carry out
Implemented in the ASE-Progol system
Learning generates hypotheses
Being studied by Ali Hafiz (PhD)
Idea: use machine learning to guide experimentation
using a real robot geneticist in a cyclic process
Aims of current project: determine the function of genes
Cost savings of 2 to 4 times over alternatives
Upcoming Nature article
FUTURE DIRECTIONS FOR MACHINE LEARNING IN BIOINFORMATICS
In-silico modelling of complete organisms
Representation and reasoning at all levels
From patient to the molecule
Probabalistic models
For more complex biological processes
Such as biochemical pathways
BIOCHEMICAL PATHWAYS
1/120th of a biochemical network