BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//chikkutakku.com//RDFCal 1.0//EN
X-WR-CALDESC:GoogleカレンダーやiCalendar形式情報を共有シェ
 アしましょう。近所のイベントから全国のイベントま
 で今日のイベント検索やスケジュールを決めるならち
 っくたっく
X-WR-CALNAME:ちっくたっく
X-WR-TIMEZONE:UTC
BEGIN:VEVENT
SUMMARY:School of IT seminar slot (compulsory if scheduled)
DTSTART;VALUE=DATE-TIME:20260430T110000Z
DTEND;VALUE=DATE-TIME:20260430T115000Z
UID:298665319401
DESCRIPTION:Title: Do Drug-Response Prediction Models Actually Use Molecu
 lar Features? A Systematic Benchmark and Ablation StudyAbstract: Numerous
  deep learning models have been proposed for drug-response prediction — 
 forecasting gene expression changes following drug treatment. While model 
 architectures have grown increasingly complex\, the lack of a unified eval
 uation framework has made fair cross-model comparison difficult\, and the 
 actual contribution of molecular features remains poorly understood. This
  study introduces a standardised evaluation framework with unified data sp
 lits\, cross-validation\, and metrics to systematically benchmark nine pub
 lished models. The framework centres on two key components: ablation via z
 ero-replacement and random shuffling of molecular feature inputs to quanti
 fy their true contribution\, and an MLP baseline using only cell-line feat
 ures to probe the relationship between architectural complexity and perfor
 mance. Preliminary results show that a simple three-layer MLP (DEG PCC 
 ≈ 0.637) matches or exceeds most complex models\, while several models s
 how little to no performance degradation when molecular features are remov
 ed. These findings highlight a widespread underutilisation of molecular in
 formation in current approaches and offer a reusable benchmark for guiding
  future model development in this field.Biography:Jinming Bai is a PhD can
 didate in the Department of Human Biology\, Faculty of Health Sciences\, U
 niversity of Cape Town\, having previously completed his MSc at UCT. His r
 esearch\, supervised by Prof Sharon Prince (cancer biology) and co-supervi
 sed by Prof Geoff Nitschke (computer science)\, applies machine learning t
 o predict post-treatment gene expression changes for drug repurposing and 
 targeted cancer therapy.
LOCATION:John Day LT1
END:VEVENT
END:VCALENDAR