[論文レビュー] Understanding Biology in the Age of Artificial Intelligence

Elsa Lawrence, A. R. Elshazly+7 more|arXiv (Cornell University)|Mar 6, 2024

Genetics, Bioinformatics, and Biomedical Research被引用数 9

ひとこと要約

The paper analyzes how machine learning reshapes biological understanding through an epistemological lens, proposing principles to guide ML design and application in biology, with focus on protein structure prediction and single-cell RNA sequencing.

ABSTRACT

Modern life sciences research is increasingly relying on artificial intelligence approaches to model biological systems, primarily centered around the use of machine learning (ML) models. Although ML is undeniably useful for identifying patterns in large, complex data sets, its widespread application in biological sciences represents a significant deviation from traditional methods of scientific inquiry. As such, the interplay between these models and scientific understanding in biology is a topic with important implications for the future of scientific research, yet it is a subject that has received little attention. Here, we draw from an epistemological toolkit to contextualize recent applications of ML in biological sciences under modern philosophical theories of understanding, identifying general principles that can guide the design and application of ML systems to model biological phenomena and advance scientific knowledge. We propose that conceptions of scientific understanding as information compression, qualitative intelligibility, and dependency relation modelling provide a useful framework for interpreting ML-mediated understanding of biological systems. Through a detailed analysis of two key application areas of ML in modern biological research - protein structure prediction and single cell RNA-sequencing - we explore how these features have thus far enabled ML systems to advance scientific understanding of their target phenomena, how they may guide the development of future ML models, and the key obstacles that remain in preventing ML from achieving its potential as a tool for biological discovery. Consideration of the epistemological features of ML applications in biology will improve the prospects of these methods to solve important problems and advance scientific understanding of living systems.

研究の動機と目的

• Motivate a philosophical and epistemological examination of ML in modern biology.
• Propose a framework for understanding ML-mediated biological insights using information compression, qualitative intelligibility, and dependency modelling.
• Analyze how ML applications in protein structure prediction and single-cell RNA sequencing affect scientific understanding.
• Identify obstacles that prevent ML from fully achieving biological discovery.
• Suggest design principles to improve future ML models in biology.

提案手法

• Survey existing ML applications in biology through an epistemological toolkit.
• Apply modern theories of scientific understanding to interpret ML-driven insights.
• Develop a framework based on information compression, intelligibility, and dependency modelling to assess ML outputs.

実験結果

リサーチクエスチョン

• RQ1 How can concepts of information compression, qualitative intelligibility, and dependency relations illuminate ML-mediated biological understanding?
• RQ2 In what ways do current ML applications in biology (e.g., protein structure prediction, single-cell RNA sequencing) advance or limit scientific understanding?
• RQ3 What design considerations should guide future ML models to enhance biological discovery?

主な発見

• ML applications in biology can be interpreted through information compression, qualitative intelligibility, and dependency modelling to understand their scientific value.
• Protein structure prediction and single-cell RNA sequencing serve as focal cases to examine how ML advances understanding.
• There are remaining obstacles that prevent ML from fully achieving its potential as a tool for biological discovery.
• The epistemological features of ML applications can guide future model development and improve problem-solving in biology.