Trust, Reproducibility, and Progress: The Role of Assessment & Benchmarking in Computational Biology

Workshop Overview

This workshop will explore the critical role of assessment and benchmarking in computational biology, emphasizing its importance in ensuring reproducibility, robustness, and innovation across disciplines such as genomics, structural biology, and biomedical data science. With the increasing reliance on AI and deep learning, the need for reliable assessment has never been greater. Participants will gain insights into best practices for designing assessments, selecting gold-standard datasets, and establishing meaningful performance metrics.

Key topics:

• Community-driven assessment and benchmarking initiatives to identify strengths and limitations of current methods
• Case studies showcasing the impact of assessment and benchmarking on research funding and regulatory decisions
• Addressing gaps in standards for reproducibility and performance evaluation in AI-driven research
• Lessons from collaborative competitions like CASP, CAGI, CAFA, and DREAM Challenges

Organizers

• Gaia Andreoletti
• Serghei Mangul
• Mark Robinson
• Predrag Radivojac
• Steven E. Brenner

Contact Information

• Gaia Andreoletti: gaia dot andreoletti at sagebase dot org
• Serghei Mangul: serghei dot mangul at sagebase dot org
• Predrag Radivojac: p dot radivojac at northeastern dot edu
• Steven E. Brenner: brenner at compbio dot berkeley dot edu