機械学習と遺伝的アルゴリズムを用いたエンハンサーの同定とモデル化

2017 Fiscal Year Annual Research Report

• Project/Area Number: 17F17797
• Research Institution: The University of Tokyo
• Principal Investigator: 中井 謙太 東京大学, 医科学研究所, 教授 (60217643)
• Co-Investigator(Kenkyū-buntansha): BERTHIER VINCENT 東京大学, 医科学研究所, 外国人特別研究員
• Project Period (FY): 2017-11-10 – 2020-03-31
• Keywords: Optimization / Genetic Algorithm / DNA / motif finding

Outline of Annual Research Achievements

In the past four month since I've arrived in the Functional Analysis in Silico Lab (Nakai Lab), I've dedicated my research to familiarize myself with the domain by developping a new algorithm able to identify target motifs in DNA. The methos is novel, and has now been tested on artificial data. At the moment, I am in the process of evaluating and tweaking the algorithm on a community approved testbed.
If the results there are as good as hoped, I will write a paper in the next couple of months.

Current Status of Research Progress

2: Research has progressed on the whole more than it was originally planned.

Reason

The first goals where well defined and I quickly got access to everything I needed to allow me to reach the objectives.

Strategy for Future Research Activity

In this year, the goal will be to improve the techniques already used, first to increase their performances, and second to allow them to be more robust, making them usable on a wider domain. This will necessitate a lot of computing power, requiring a very good working station with a high number of efficient CPU cores. Speeding up this process could require access to a computing cluster.