Agentic AI for Biomedical Discovery

Erscheinungsjahr 2027
ISBN: 978-3-032-46499-6
Verlag: Springer

Buch, Englisch, Format (B × H): 155 mm x 235 mm

Reihe: Bio-IT and AI

This book is a guide to agentic AI in biomedical research: systems built around large language models that plan analyses, call software tools and databases, write and run code, and revise their own work. It covers how these systems are built, through architectures, knowledge integration, tool use, hypothesis refinement, long multi-step analyses, and verification and provenance; how they are applied in genomics, single-cell and spatial transcriptomics, proteomics, structural biology, drug discovery, disease mechanisms, clinical research and population genomics; and the human–AI collaboration, trustworthiness and governance questions that surround them.
Each chapter builds up a domain’s own methods before surveying the agents built on top of them, and reports what each system was evaluated on and what can fairly be compared with what. A closing Perspective in every chapter then states the judgment the evidence supports, so readers can separate benchmark scores, expert preferences and validated findings.
The book is written for researchers who build these systems and for biomedical scientists who need to decide what their results can be trusted to show.

Zielgruppe

Research

Autoren/Hrsg.

Wang, Haohan

Fachgebiete

• Mathematik | Informatik EDV | Informatik Angewandte Informatik Computeranwendungen in Wissenschaft & Technologie
• Mathematik | Informatik EDV | Informatik Angewandte Informatik Bioinformatik
• Technische Wissenschaften Technik Allgemein Computeranwendungen in der Technik
• Mathematik | Informatik EDV | Informatik Informatik Künstliche Intelligenz
• Naturwissenschaften Biowissenschaften Angewandte Biologie Bioinformatik

Inhaltsverzeichnis

.- Part I: Foundations.

.- Chapter 1: The Promise of Agentic AI in Biomedical Discovery.

.- Chapter 2: Multi-Agent Architectures for Biology.

.- Part II: Core Capabilities.

.- Chapter 3: Knowledge Integration and Reasoning.

.- Chapter 4: Tool Use and Programmatic Control.

.- Chapter 5: Iterative Hypothesis Refinement.

.- Chapter 6: Multi-Step Analysis and Decision Making.

.- Chapter 7: Verification, Quality Control, and Provenance.

.- Part III: Domain Applications.

.- Chapter 8: Genomics and Gene Regulation.

.- Chapter 9: Transcriptomics: Single-Cell and Spatial Analysis.

.- Chapter 10: Proteomics and Multi-Omics Integration.

Chapter 11: Structural Biology and Protein Design.

.- Chapter 12: Drug Discovery and Molecular Therapeutics.

.- Chapter 13: Disease Mechanism Discover.

.- Chapter 14: Clinical Data and Translational Research.

.- Chapter 15: Population Genomics and Evolutionary Biology.

.- Part IV: Practice and Perspective.

.- Chapter 16: Human-AI Collaboration in Biomedical Research.

.- Chapter 17: Trustworthy Agentic Systems.

.- Chapter 18: Responsible Development and Governance.

.- Chapter 19: The Future of AI-Accelerated Science.

Über Autor(innen)

Haohan Wang is an assistant professor in the School of Information Sciences at the University of Illinois Urbana-Champaign, where he leads the DREAM Lab, with affiliated appointments in the Siebel School of Computing and Data Science, the Carl R. Woese Institute for Genomic Biology, the Beckman Institute for Advanced Science and Technology, and the National Center for Supercomputing Applications. His research develops AI methods for scientific discovery, spanning agentic AI systems that act as virtual bioinformaticians, computational methods for the genetic basis of complex diseases such as Alzheimer’s disease and cancer, and trustworthy and secure machine learning. He received his PhD (2021) and MS (2015) from the Language Technologies Institute at Carnegie Mellon University, and was a postdoctoral researcher in the Machine Learning Department at Carnegie Mellon before joining Illinois in 2022.