BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Proteomizer &amp\; GhostBuster: AI Tools for Proteomic Inference a
 nd Ghost Gene Annotation - Giulio Deangeli
DTSTART:20250716T160000Z
DTEND:20250716T170000Z
UID:TALK233938@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:Despite the rapid growth of multiomic datasets\, our ability t
 o interpret and integrate transcriptomic\, proteomic\, and regulatory info
 rmation remains limited by both biological complexity and systemic biases 
 in data and literature. In this talk\, I will present two complementary ma
 chine learning frameworks—Proteomizer and GhostBuster—that address the
 se challenges from distinct but synergistic angles.\n\nProteomizer is a de
 ep learning platform that predicts proteomic landscapes from transcriptomi
 c and miRNomic profiles\, achieving state-of-the-art accuracy (r = 0.68) o
 n over 8\,600 matched samples. Beyond prediction\, Proteomizer enhances di
 fferential expression analysis and enables mechanistic insights through ex
 plainable AI\, revealing regulatory interactions that underlie transcript-
 protein discrepancies.\n\nGhostBuster\, on the other hand\, tackles a diff
 erent but equally critical issue: literature bias in gene annotation. Many
  human genes remain understudied due to sociological dynamics that skew re
 search focus. GhostBuster is the first ML framework explicitly designed to
  mitigate this bias\, using unbiased datasets (e.g.\, TCGA\, LINCS) to unc
 over novel gene functions\, disease associations\, and pathway memberships
 . It demonstrates that models trained on less-biased data are significantl
 y more effective at identifying emerging biological knowledge\, particular
 ly for "ghost genes".\n\nTogether\, these tools exemplify a new generation
  of interpretable\, bias-aware machine learning approaches that not only i
 mprove predictive performance but also expand our capacity to generate bio
 logically meaningful hypotheses—especially for the vast uncharted region
 s of the human genome.
LOCATION:Lecture Theatre 2
END:VEVENT
END:VCALENDAR