Menu: (Choose Your Destination) BCB Courses BCB Home Computing Laboratory Curriculum & Training Faculty Directory Faculty Research ISU Bioinformatics Sites National Science Foundation Training Grant Overview of BCB Research Participating Departments Prospective Students Site Index Student Resources University Community Upcoming Seminars

BCB 691 - Faculty Seminar Fall 2005

Course Details and Abstracts below.

1)  to introduce new BCB students to examples of research rotation/exploration opportunities available in BCB research groups
2)  to provide all BCB students with an overview of active research areas in Bioinformatics and Computational Biology at ISU

Course Website: http://www.bcb.iastate.edu/courses/BCB691-F2005.html

Requirements:  Attendance and participation.  Each student is allowed one "excused absence" (missed due to conference attendance, illness, family obligation, etc.) and one "unexcused absence" (rather eat lunch, whatever...). Students who exceed these limits and wish to pass the course will be expected to perform makeup work (see class schedule for Dec. 2 and 9).

Questions:  contact Srinivas Aluru and Chris Tuggle , instructors aluru@iastate.edu and cktuggle@iastate.edu
                               3227 Coover, 4-3539; 2255 Kildee, 4-4252

Assembling the Tree of Life

The Tree of Life connects over 1.75 million known life forms on earth through a tree like network of evolutionary relationships. A complete description of this tree would provide biologists with a predictive power similar to the one that chemists have from the Periodic Table of Elements. Thus knowing the Tree of Life, or large parts of it, would result in an enormous benefit to science and society. For example human health could be improved and the frontiers in agriculture and comparative biological could be pushed forward.

Unfortunately, only very small parts of the Tree of Life could be assembled so far. Systematic biologists are faced with the problem to reassemble the Tree of Life from the genetic information of over 1.75 million known species. In analogy, this task is similar to the task of assembling an alien starship from its pieces without having any documentation. However, recent interdisciplinary developments of powerful computational tools and the availability of major new data sources might allow systematic biologists to assemble larger parts of the Tree of Life within the next two decades.

In this presentation I will pinpoint the frontier of computational problems for constructing the Tree of Life.

Evolutionary Genomics: From Gene to Genome

Abstract

Our long-term research goal is to study evolutionary/comparative genomics, population genetics and computational biology. My lab is developing research projects with a combination of statistical/computational methods, software engineering, large-scale multi-layer data analysis, and experimental work. These research projects include:

Evolutionary Functional Genomics
- Statistical framework for Gene family Evolution
- Integrated Software System

- Evidence for the Association between Site-Specific Rate Shifts and Changes in Function after Gene Duplication Comparative and Evolutionary Genomics
- Vertebrate genome duplication & origin of human gene family hierarchy
- Algorithms for ancestral gene order inference and comparative genome mapping
- Whole-genome phylogenetic analysis based on gene (family) content

Multi-Layer Genome Data Exploration and Gene Network Evolution
- Statistical framework for expression profile evolution
- Evolution of repeat elements, gene regulation and motif detecting
- Genetic buffering, duplicate genes and network complexity

Human Population Genetics and Primate Evolution
- Gene expression evolution in humans and chimpanzees
- Interplay between species evolution and population genetics

Computational Gene Prediction -- Methods and Applications

Abstract:

Modern biology research is characterized by the ability to study questions from a genome-wide perspective. Whereas only a decade ago a research project would typically focus on a single gene or pathway, it is now possible to view and evaluate the same genes and pathways in the context of all the genes of an organism, mapped onto the chromosomes that constitute the species' entire genetic blueprint. Of course, these possibilities require prior correct identification and annotation of all the genes, a challenging problem that has not been entirely solved. Whereas genomic DNA sequencing and assembly is a mostly technological process, gene annotation is largely computational, involving both statistically based prediction methods and integration of various sources of experimental and knowledge-based evidence.

In this talk we give an overview of the computational gene prediction field. Thereby, we focus on similarity-based methods. Finally, we give some results of predicting genes on the newly sequenced chimpanzee genome.

Title: Daily bioinformatics - sampling the use of computers in everyday biological research

Abstract:

Development in bioinformatics has been steadily expanding in the past several years. Many of the hard computational problems in bioinformatics such as genome assembly or microarray analysis have drawn much of bioinformaticists' attention, resulted in powerful and complex software tools for specialized jobs. In this talk, however, I am presenting an alternative perspective, one that suggests even simple, daily bioinformatics can benefit biologists a lot.

New Methods for Simulating, Refining, and Modeling Supermolecular Complexes at Multi-resolution and Multi-length Scales

Abstract:

A set of new computational methods has been developed for simulating, refining, and modeling supermolecular complexes at multi-resolution and multi-length scales. On the resolution scale, quantized elastic deformational model (QEDM) was designed to reliably describe large-scale protein motions in the absence of aminoacid sequence and atomic coordinates. QEDM yields an accurate description of protein dynamics over a wide range of resolutions even as low as 30 Å. On the length scale, substructure synthesis method (SSM) was developed to derive the motions of a given structure as a collection of those of an assemblage of substructures. SSM was applied to F-actin, a typical filamentous system in cells. The results demonstrated that SSM is capable of scaling the simulation of atomic motions of molecular complexes to a macroscopic length scale. The QEDM and SSM methods have been successfully applied to assisting structural refinement against cryo-EM and fibre diffraction data, respectively. The results demonstrated that, under the scheme of harmonic modal analysis, structural refinement for seemingly remote experimental techniques can be unified for systems that involve large-scale conformational flexibility. In order to improve one’s ability to interpret low- to intermediate- resolution density maps, a series of computational methods have been developed. They are methods like sheetminer and sheettracer that are capable of extracting secondary structural features, and methods that can determinate protein topology merely based on information of secondary structural skeletons. Methods has also been developed for protein folding assisted by SAXS data which carries hope to significantly improve the effective resolution of SAXS technique. Results have shown that SAXS data carry necessary information that is sufficient to derive overall fold of proteins. Such methods will bridge the gap between cryo-EM and xray crystallography, in which the small, soluble, and noncrystalizable proteins can be studied.

Protein Networks

Abstract:

New types of analyses are necessary to reveal organizational principles of protein interaction networks for investigating the details of the functional and regulatory clusters of proteins. Eigenmode analysis of the entire connectivity matrix yields both a global and a detailed view of the network by which individual functional clusters can be probed. Structural clustering of the protein interactome network reveals functionalities of individual proteins, many having previously unknown functions. Simulations of protein motions are feasible using simple network representations reveal their most important functional motions through eigenmode analyses. We are beginning to construct larger structures from the interacting proteins for which we can perform large-scale simulations. Consideration of the effects of binding on protein motions seems to indicate that different proteins could act as repressors or enhancers of functional motions.

Assembly and Alignment of Genomic DNA Sequences

Abstract

I will present our recent work on genome assembly and multiple alignment of syntenic genomic sequences. The PCAP program builds a genome assembly almost free of global misjoins from millions of short sequences. PCAP has been used in chicken and chimpanzee genome projects at Washington University in St. Louis. The MAP2 program computes an ordered list of short conserved regions (exon and regulatory regions) between long, different regions (intron and intergenic regions).

Srinivas Aluru
Dept. of Electrical and Computer Engineering

How to do sequence alignments on parallel computers?

Abstract

In this talk, I will present algorithms for carrying out sequence alignments using parallel computers. More specifically, given two sequences of length m and n, and p processors, we would like to do sequence alignment in O(mn/p) time using O((m+n)/p) memory per processor. I will show how the techniques presented here can be used to perform in parallel, the time consuming human vs. mouse syntenic alignment Prof. Huang discussed in the previous seminar.

November 11

Stephen Willson
Mathematics Department

Building Supertrees Using Distances

Abstract:

Suppose that a family of rooted phylogenetic trees with different sets of leaves is given. A supertree for the family would be a single rooted tree T whose leaf set is the union of all the input leaf sets, such that the branching information in T corresponds to the branching information in all the input trees. This talk gives an overview of some methods for finding supertrees. It focuses on a polynomial-time method BUILD-WITH-DISTANCES that makes essential use of distance information provided on the input trees. When a supertree containing also the distance information exists, then the method produces a supertree T. This supertree often shows increased resolution over the trees found by methods that utilize only the topology of the input trees. When no strict supertree exists because the input trees are incompatible, an extension of the method still produces a tree with interesting properties.

November 18

Dan Nettleton
Statistics Department

Using P-Values for the Planning and Analysis of Microarray Experiments

Abstract: 

Microarray experiments to identify genes that change expression across multiple conditions can be used to gain clues about gene function.  Many analysis strategies involve obtaining a p-value for a test of differential expression for each of thousands of genes.  I will discuss an intuitively appealing method, originally proposed as an iterative algorithm by Mosig et al. (2001, Genetics 157, 1683-1698), for estimating the total number of differentially expressed genes and the False Discovery Rate (FDR) associated with any threshold for statistical significance.  I will characterize the limit of the iterative algorithm and describe how the estimator can be computed directly without iteration.  I will compare the performance of the resulting simple estimator with other procedures for estimating the number of true null hypotheses from a collection of observed p-values.  I will conclude with a discussion of a new method that can provide information about power and sample size for a future experiment based on p-values from a pilot experiment.

Nov. 25 - Thanksgiving Break

December 2

Dr. Vasant Honavar
Department of Computer Science
Artificial Intelligence Research Laboratory
Center for Computational Intelligence, Learning, and Discovery
Bioinformatics and Computational Biology Graduate Program

Acknowledgements: The  research on INDUS has been carried out in collaboration with members of the ISU Artificial Intelligence Research Laboratory.  Research in bioinformatics applications has been carried out in collaboration with Drena Dobbs, Robert Jernigan, Heather Greenlee and several other members of the ISU Bioinformatics and Computational Biology Program. This research has been supported in part by Iowa State University and grants from the National Science Foundation (IIS 0219699),  the National Institutes of Health (GM 0066387), and the US Department of Agriculture.

December 9

Julie Dickerson
Electrical and Computer Engineering Department

The talk will be two talks:

Julie Dickerson: Visualization of Biological Networks for Analysis

Graph models are being proposed for modeling and visualizing network
interactions. However as graphs grow larger to show complex interactions
between pathways, they begin to become impossible to interpret. This talk
will cover how we are working to address this problem.


Pan Du: Learning Time-Dependent Gene Regulatory Networks

This work integrates multi-scale clustering and short-time correlation to
estimate regulatory networks with different time resolutions and degrees
of co-regulation. The algorithm was evaluated using yeast cell cycle data.
The results give the networks at different levels of detail, and reflect
most interactions previously identified by genome-wide location analysis.

Fall 2003 Speakers: 10' Rotation Presentations:  Carpenter, Dobbs, Miller, Minion, Shoemaker, Tuggle  40' Speakers: Aluru, Brendel, Dobbs, Dorman, Honavar, Peccoud, Sannier, Smiley, Song, Travesset, Voytas, Wurtele

Fall 2002 Speakers:  Andreotti, Bogdanove, Ashlock, Carpenter, Cook, Gu, Hong, Levine, Mayfield, Miller, Wise

Fall 2001 Speakers: Adams, Aluru, Bhattacharyya, Dickerson, Dorman, Huang, Nettleton, Voytas, Wise, Wurtele

Fall 2000 Speakers: Aluru, Ashlock, Dickerson, Eulenstein, Ho, Honzatko, Honavar, Miller, Stern, Wurtele

URL:
Copyright© 2005, Iowa State University, all rights reserved.
Please direct corrections, suggestions, and comments to bcb@iastate.edu.
Last Modified: