Do Science! with Prof. Krzysztof Fidelis "Protein structure prediction in CASP: From homology modeling to solving the protein folding problem; prospects for the future", 25.10.2022(Tuesday), room 003, IIMCB

2 views
Skip to first unread message

Marcin Magnus

unread,
Oct 24, 2022, 10:57:24 AM10/24/22
to do-science...@googlegroups.com, All IMol
Reminder ;) TOMORROW!

Dear All,

We're back offline! Join us for the old, good, classic Do Science @IIMCB (https://doscience.iimcb.gov.pl). This time Krzysztof will explain the latest progress in protein structure prediction and prospects for the future. More on Krzysztof below. Everyone is welcome (students, postdocs, PIs)!

The program of the event is as follows:

TUESDAY, OCTOBER 25, 2022 AT 6 PM
18:00 Lecture @ Do Science! Science Club [IIMCB room 003]
Protein structure prediction in CASP: From homology modeling to solving the protein folding problem; prospects for the future.

19:00 Discussion: How to succeed in science? ;-)
20:00 Beer/coffee @Jeleń na Rykowisku [Grójecka 118]


Krzysztof Fidelis is Co-Founder (with John Moult, 1994) of the Critical Assessment of Structure Prediction initiative (CASP) to evaluate modeling methods in structural biology. He also heads the Protein Structure Prediction Center, first at Lawrence Livermore National Laboratory (1994-2005), and then at UC Davis (2005-present). CASP is designed to advance the modeling of protein structure by conducting community-wide assessments that determine the strengths and weaknesses of current methods and so foster progress. The Prediction Center provides the infrastructure for these evaluations. Currently, the fifteenth biennial assessment is underway. Approximately one hundred research groups world-wide participate, with over 67,000 submissions on approximately one hundred protein modeling targets submitted in the last round.

To help navigate through structure similarities in proteins, Krzysztof’s group introduced the local structure descriptors, defined as amino acid-attached structure context. Their mid-level range of granularity allows for characterizing structure in “modules” that are small and common enough to be found in many proteins. The formalism was applied in several areas, including identification of structure similarity, fold recognition, enzyme-ligand interactions, and structure-function relationships.

Although the primary thrust of his research is in development of methods for structure modeling assessment, he has participated in several collaborations in applied modeling, for example producing the structural models for the Rad1 cell cycle checkpoint proteins, later confirmed by experiment. 

Krzysztof’s group has also used genome-wide yeast binding-site and microarray expression data to show the combinatorial nature of gene regulation. These results provided specific data on how binding sites may be combined to allow a large number of expression outcomes using relatively few transcription factors.

https://www.facebook.com/events/547035810758623

See you soon!

DoScience! Team



Reply all
Reply to author
Forward
0 new messages