cis logo

2015 IEEE Conference on
Computational Intelligence in Bioinformatics and Computational Biology

Crowne Plaza Niagara Falls - Fallsview, Niagara Falls, Canada August 12-15, 2015

Tutorial Schedule and Links to Materials

ieee logo
 

Note: tutorial materials will be linked into this page on the day of the tutorials, Wednesday, August 12th.

Schedule

All tutorials are on Wednesday, August 12th in the Niagara Room

8:30AM-10:00AM
A Survey of Representations for Applying Evolutionary Computation to Bioinformatics
Daniel Ashlock

    Slides
10:30AM-12:00PM
Bioinformatics on the Epigenome
Wendy Ashlock

    Slides
1:30PM-3:00PM
Applications of Bayesian Model Averaging in Personalized Medicine and Systems Biology
Ka Yee Yeung

    Slides
Google docs with info
3:30PM-5:00PM
The Identification of cis-Regulatory Elements: A Tutorial From the Machine-Learning Perspective
Yifeng Li
    (slides not yet available)


 

Abstracts

d.ashlock   8:30AM-10:00AM
A Survey of Representations for Applying Evolutionary Computation to Bioinformatics
Daniel Ashlock

Representation is a central issue in evolutionary computation and the study of representation opens any options for solving problems in bioinformatics. The no free lunch theorem demonstrates that here is no intrinsic advantage in a particular algorithm when considered against complete spaces f problems. The corollary is that algorithms should be fitted to the problems they are solving. hoice of representation is the primary point in the design of an evolutionary algorithm where he designer can incorporate domain knowledge and, in effect, choose the adaptive landscape he is earching. Bioinformatic is a field rich in available domain knowledge and so ripe for representational nnovation. This tutorial will introduce a broad variety of bioinformatics-relevant representations ith comments on their adaptive landscapes and domains of application. These will included novel enerative representations, ordered gene representations, state conditioned representations and gen- ral modifications that can be made to a variety of representations. Familiarity with evolutionary lgorithms is assumed, but no other special knowledge is required.
 
w.ashlock   10:30AM-12:00PM
Bioinformatics on the Epigenome
Wendy Ashlock

When the Human Genome Project was first proposed in 1984, there was great hope that its results would lead to cures for scores of diseases, especially cancers. But, it was completed twelve years ago. Why haven't we cured cancer yet? The answer is that human genetics is much more complex than anticipated. Diseases cannot be cured by just finding the relevant gene and ``fixing'' it. One of the unexpected complications is the existence of the epigenome. Epigenetics involves heritable changes to the phenotype that are not changes to the genotype. The epigenome consists of the chemical tags on the DNA that result in epigenetic changes. This tutorial will introduce the biology behind epigenetic change, including methylation, histone modification, and a type of epigenetics unique to the binuclear single-celled ciliates. Publically available data sources for bioinformatic data will be presented, and an example project using computational intelligence will be described.
 
w.ashlock   1:30PM-3:00PM
Applications of Bayesian Model Averaging in Personalized Medicine and Systems Biology
Ka Yee Yeung

Modern biology is very much a data rich science. In this tutorial, we will introduce Bayesian Model Averaging (BMA) methods and its applications in computational biology. BMA is a multivariate variable selection method that accounts for model uncertainty. We will also introduce big biological data and illustrate the applications of BMA using these big data. In particular, we will focus on two applications: personalized medicine and systems biology.

Personalized (or precision) medicine aims to customize individual patients’ treatment by accounting for genetic variations between individuals. With the Precision Medicine Initiative announced by President Obama in the State of the Union Address in January 2015, this area has received substantial media attention. We will discuss how BMA can be used in a supervised learning framework to identify gene signatures predictive of clinical outcomes. We will use our previous work on chronic myeloid leukemia as a case study.

Systems biology is the study of complex interactions between biological entities. Networks are ubiquitous in science. We will discuss how BMA can be used in a regression framework to identify gene-to-gene influences using big biological data. We will use our previous work on DREAM time series data as a case study.
 
li   3:30PM-5:00PM
The Identification of cis-Regulatory Elements: A Tutorial From the Machine-Learning Perspective
Yifeng Li

The majority of the human genome consists of non-coding regions that have been called junk DNA. However, recent studies have unveiled that the sequences contain cis-regulatory elements, such as promoters, enhancers, silencers, insulators, etc. These regulatory elements can play crucial roles in controlling the expression of genes in specic cell types, conditions, and developmental stages. Disruption of these regions could contribute to phenotype changes. Precisely identifying regulatory elements is key to deciphering the mechanisms underlying transcriptional regulation. Cis-regulatory events are complex processes that involve chromatin accessibility, transcription factor binding, DNA methylation, histone modications, and the interactions be- tween them. The development of next-generation sequencing techniques has enabled us to capture these genomic features in depth. Clinical genetics makes the detection of these regions a priority. However, the complexity of cis-regulatory events and the deluge of sequencing data require accurate and e cient computational approaches, in particular, machine learning techniques. In this tutorial, we describe machine learning approaches for predicting transcription factor binding sites, enhancers, and promoters, primarily driven by next-generation sequencing data. Data sources are provided in order to facilitate testing of novel methods. The purpose of this tutorial is to attract computational experts and data scientists to advance this field.