TODAY at 2pm NSH 3305: Talk on Collaboration ideas in AI for Science by Professor Koyama, Keio University, Japan

[Artur Dubrawski]

Team,

Please attend this Auton-lab hosted presentation:

Professor Takahiko Koyama, visiting from Keio University, Japan, will give
a talk at 2pm on Thursday March 13, in NSH 3305.

Zoom: https://cmu.zoom.us/j/93727472083?pwd=yfVKqQ3TeH9HGnPJoGxuVGa1tYu3Sx.1
Meeting ID: 937 2747 2083
Passcode: 955218

Title:
Collaboration ideas in AI for Science

Abstract:
The establishment of a new collaboration between Keio University and
Carnegie Mellon University marks an important advancement in artificial
intelligence research, with a special focus on AI applications in
scientific discovery. This presentation will outline our emerging
partnership and highlight potential areas for joint research initiatives.
Our laboratory at Keio University Bio2Q brings specialized expertise in
microbiome research and quantum computing to this collaboration. Among the
key research directions is the development of genome language models, which
represent a promising frontier in computational biology as well as
genomics. With employment of genome language model and protein language
model, we are conducting in-depth annotations of gut microbiome genes,
which have been naively annotated using sequence similarities. Furthermore,
we are developing a transformer based AI model named gut MPT to predict and
identify unknown metabolites—often referred to as "dark matter
metabolites"—which remain largely uncharacterized despite their potential
significance in human health. In another frontier, we are paving a way to
build quantum algorithms for medicine and biology realms. Through this
collaboration, we aim to leverage complementary strengths across
institutions to address these complex scientific challenges and advance the
frontiers of AI for scientific discovery.

Speaker's Bio:
Born in Osaka, Japan with dreams of aerospace engineering, Taka's path
changed after discovering Richard Feynman's "QED," leading to a Ph.D. in
physics from Cornell University. Inspired by the Human Genome Project, he
transitioned to life sciences through postdoctoral research at the
University of Pennsylvania, studying protein flexibility and bioinformatics.
At Takeda Pharmaceutical, he specialized in bioinformatics and
computational chemistry, developing highly specific allosteric kinase
inhibitors for cancers, and applying machine learning to high-throughput
library design. Later at IBM Watson Research Center, he led a team of
scientists and engineers for Watson for Genomics and published influential
work on SARS-CoV-2 genome variant analysis ahead of the Chinese Academy of
Science.
Today, Taka combines expertise in AI, quantum computing, and biology to
explore microbiome science, decoding metabolomic "dark matter" and
investigating virome functions through quantum algorithm developments.

Note:
We have a couple of 1-on-1 meeting slots available on Friday morning
between 11:00 and noon, please email your bids to Artur Dubrawski (
awd at cs.cmu.edu), who will allocate them on the first-come-first served
basis.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20250313/97916460/attachment.html>