BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:721c8005a9cd5db281334447f6f54a09
DTSTAMP:20260606T102015Z
SUMMARY:Title: Large Scale Machine Learning Methods for Predicting Protein 
 Functions
DESCRIPTION:Speakers: Vladimir Gligorijevic and Richard A Bonneau (Flatiron
  Institute\, New York) \nTitle: Large Scale Machine Learning Methods for 
 Predicting Protein Functions from Sequence\, Structure and Networks \nAbs
 tract:  Due to limitations of existing experimental methods for determini
 ng protein functions and the high cost of experiments\, the vast majority 
 of proteins across many organisms remain unannotated. Developing ML metho
 ds for combining large-scale genome-wide heterogeneous data to extract use
 ful protein feature representations for function prediction thus remains 
 a key problem in biology.   We will review a few of our recent deep learn
 ing-based methods for predicting function from various data types\, includ
 ing protein sequences\, structures and protein-protein interaction network
 s. We will then present our recent integrative method\, deepNF (deep netwo
 rk fusion) that integrates different protein-protein interaction networks 
 using auto-encoders to construct a shared protein feature representation i
 ndicative of protein function. In the second part of the talk\, we will fo
 cus more on integrative methods for predicting function from protein seque
 nce and structure. Lastly\, we will present a method based on graph convol
 utional networks (GCNs) which extracts features from protein contact maps 
 and sequences while learning to predict function. We will discuss the perf
 ormance of GCN in predicting functions of experimentally determined struct
 ures from PDB and how we are planning on applying this method on Rosetta-p
 redicted structures.
URL:https://www.imperial.ac.uk/events/97585/title-large-scale-machine-learn
 ing-methods-for-predicting-protein-functions/
DTSTART;TZID=Europe/London:20190110T140000
DTEND;TZID=Europe/London:20190110T150000
LOCATION:United Kingdom
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:STANDARD
DTSTART:20190110T140000
TZNAME:GMT
TZOFFSETTO:+0000
TZOFFSETFROM:+0000
END:STANDARD
END:VTIMEZONE
END:VCALENDAR