Background/Education

My name is Fadi Towfic. I received my doctorate in Bioinformatics and Computational Biology at Iowa State University. I joined Dr. Vasant Honavar's Artificial Intelligence Research Laboratory in the summer of 2007. I am also a member of the Center for Computational Intelligence, Learning, & Discovery. I earned dual Bachelor degrees in Computer Science and Biology (Genetics/Biotechnology track) from the University of Iowa with a minor in Chemistry. I have always loved to apply techniques from Computer Science to problems from Biology and Chemistry, as those sciences afforded many challenges that are similar to problems encountered in Computer Science (e.g Largest Common Subsequence problem and Sequence Alignment), yet with unique twists that make the problem less approachable through traditional means (e.g. gap insertion and gap penalties in sequence alignment algorithms). Those twists require the use of techniques from fields such as statistics or mathematics to evaluate the performance of algorithms. As an undergraduate, I worked with Dr. John Logsdon Jr. to study the evolution of multimeric proteins. It was in the Logsdon lab where I was introduced to the field of Bioinformatics and the unique opportunities presented by this emerging science. After a brief period as a full-time Systems Analyst at Principal Financial Group (awesome place to work, by the way). I was admitted as a full time graduate student to Iowa State University's Bioinformatics and Computational Biology graduate program and granted an IGERT fellowship. I happily worked in the Honavar lab and collaborated with Dr. Heather Greenlee, Dr. Drena Dobbs, Dr. Chris Tuggle and Dr. Robert Jernigan on various projects involving the modeling, searching and comparison of biological networks.

Research Interests

Biological Networks

My main research interest in bioinformatics is in the construction and comparison of biological networks (gene regulatory networks and protein-protein interaction networks). As part of my research, I have developed the Biomolecular Network Alignment toolkit. I have also contributed improvements to BioNetwork Bench to efficiently construct correlation networks from microarray data. My current work is focused on developing efficient algorithms for comparing biological networks constructed from various sources.

Protein Interface Prediction

I am interested in the prediction of protein-protein and protein-RNA/DNA interactions based on sequence and structural features. I have worked with several great partners in the Honavar lab to improve the prediction accuracy of protein-protein and protein-RNA interactions using the proteins' 3D structure.

Machine Learning

I am interested in applying and developing several machine learning algorithms/techniques to the bioinformatics problems I work on. For searching/comparing biological networks, several graph/matrix comparison techniques such as graph-kernels and tensors are highly relevant to my work in that area. For protein interface prediction/classification, several techniques for utilizing mixtures of experts, dealing with unbalanced datasets (e.g AdaBoost? ), or feature selection are critical for improving the prediction of interaction interfaces on proteins.

Resume

You may access a printable version of my CurriculumVitae? here.