BEGIN:VCALENDAR
PRODID:-//eluceo/ical//2.0/EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
UID:411e4ddb1edc2c24097b8660fe262ee3
DTSTAMP:20260312T200733Z
SUMMARY:Richard Bonneau (Flatiron): Structure and region-specific function 
 predictions of microbial proteins for microbiome research
DESCRIPTION:Biomaths (Mathematics) – EPSRC CMPH Seminar\n \nSpeakers:\nV
 ladimir Gligorijevic1 Tomasz Kosciolek2 Richard Bonneau1\n1 Center for Com
 putational Biology\, Flatiron Institute\, Simons Foundation\, New York\, U
 SA\n2 Małopolska Centre of Biotechnology\, Jagiellonian University\, Krak
 ów\, Poland\nSummary: \nThe human gut microbiome is estimated to harbor o
 ver 3 million unique protein-coding gene families. Only a fraction of them
  are experimentally annotated and therefore require computational predicti
 ons. Community-wide benchmarking efforts such as CAFA show that homology-b
 ased function annotation approaches are lacking and require more sophistic
 ated approaches. In this talk\, we will introduce deepFRI (deep learning-b
 ased functional residue identification)\, our recently proposed method bas
 ed on Graph Convolutional Networks [1]. We use protein families\, within a
  deep learning structure prediction framework\, to predict protein 3D stru
 ctures. We show that deepFRI achieves state-of-the-art accuracy in predict
 ing gene ontology terms of proteins with predicted 3D  structural models.
  We are now in a position to functionally annotate microbial genomes and m
 etagenomes with higher coverage and accuracy. We may also start addressing
  microbe-microbe and host-microbiome protein-protein interactions to deter
 mine the mechanisms of microbiota-induced immune response. An added value 
 of deepFRI is its ability to produce region-specific structure predictions
  through class activation mapping.\n[1] Gligorijevic\, V.\, Renfrew\, P.D.
 \, Kosciolek\, T.\, Leman\, J.K.\, Cho\, K.\, Vatanen\, T.\, Berenberg\, D
 .\, Taylor\, B.C.\, Fisk\, I.M.\, Xavier\, R.J. and Knight\, R.\, 2019. St
 ructure-Based Function Prediction using Graph Convolutional Networks. bioR
 xiv\, p.786236.\n 
URL:https://www.imperial.ac.uk/events/116158/structure-and-region-specific-
 function-predictions-of-microbial-proteins-for-microbiome-research/
DTSTART;TZID=Europe/London:20200227T160000
DTEND;TZID=Europe/London:20200227T180000
LOCATION:342\, Huxley Building\, South Kensington Campus\, Imperial College
  London\, London\, SW7 2AZ\, United Kingdom
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:STANDARD
DTSTART:20200227T160000
TZNAME:GMT
TZOFFSETTO:+0000
TZOFFSETFROM:+0000
END:STANDARD
END:VTIMEZONE
END:VCALENDAR