Ali Sharifi Zarchi

Assistant Professor

Integrating Algorithmic Methods and Artificial Intelligence: A New Paradigm in Bioinformatics

  Halls department, Hall 2
  Wednesday, 27 December 2017
  15:30 - 16:30


During past decades, a major framework of bioinformatics has been an integration between algorithmic and statistical methods. For example, an analysis pipeline for Next Generation Sequencing (NGS) data might consist of some algorithmic pre-processing methods, such as alignment of the reads to a reference genome, followed by statistical post-processing methods, such as genomic variations calling. A major challenge for this framework has been to find statistical parameters and thresholds in order to accurately classify between positive and negative cases. These parameters are usually fine-tuned on some specific data, and might not work for data of other labs, organisms, cell types, sequencing coverages or platforms.
In this talk, we review some examples of using deep neural networks for solving some bioinformatics problems. For instance, we show how the identity of a cell can be accurately predicted from its gene expression profile. These methods do not require fine-tuning some parameters and can work for a broad range of biological datasets. We suggest using artificial intelligence techniques in combination with algorithmic methods can be a new paradigm for solving biological problems.


Ali Sharifi-Zarchi received his Bachelor and Master degrees from the Sharif University of Technology in Computer Engineering, and his Ph.D. degree in bioinformatics form the Institute of Biochemistry and Biophysics in the University of Tehran under the supervision of Dr. Mehdi Sadeghi and Dr. Hamid Pezeshk. Now he is doing bioinformatics research at Royan Institute and is an assistant prof. of bioinformatics at the department of computer engineering at the Sharif University of Technology.