BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//euroscipy-2025//speaker//EVNBNX
BEGIN:VTIMEZONE
TZID:Europe/Warsaw
BEGIN:DAYLIGHT
DTSTART:20240821T000000
TZNAME:CEST
TZOFFSETFROM:+0200
TZOFFSETTO:+0200
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20241027T030000
RDATE:20251026T030000
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20250330T030000
RDATE:20260329T030000
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
SUMMARY:Efficient and accurate models for peptide function prediction - Pi
 otr Ludynia
DTSTART;TZID=Europe/Warsaw:20250821T103000
DTEND;TZID=Europe/Warsaw:20250821T110000
DTSTAMP:20260908T131609Z
UID:pretalx-euroscipy-2025-CTCYRB@pretalx.com
DESCRIPTION:Peptides are small proteins\, regularing many important biolog
 ical processes. They have significant therapeutic potential\, thanks to th
 eir properties\, e.g. microbial\, antiviral\, or anticancer. \nIn particul
 ar\, they offer a promising alternative to traditional antibiotics\, addre
 ssing the growing crisis of drug resistance.\nAccurately predicting peptid
 e properties is essential for drug discovery\, and recent research has exp
 lored deep learning approaches such as graph neural networks\, protein lan
 guage models\, and multimodal ensembles.\nHowever\, these methods are ofte
 n overly complex and lack scalability. They are also brittle and their per
 formance breaks down on new datasets or tasks.\nWe propose to use molecula
 r fingerprints for this task. They are established feature extraction algo
 rithms from chemoinformatics\, primarily applied on small molecules. \nWe 
 show that they obtain state-of-the-art results on peptide function predict
 ion and can efficiently vectorize larger biomolecules.\nThis approach is s
 imple\, fast\, and accurate. We comprehensively measure its robustness on 
 6 benchmarks and 126 datasets. This unlocks a novel venue in chemoinformat
 ics-based approaches for peptide-based drug design.
LOCATION:Room 1.20 (Ground Floor\, Shannon)
URL:https://pretalx.com/euroscipy-2025/talk/CTCYRB/
END:VEVENT
END:VCALENDAR
