Machine learning in crystallography and structural science.

Area of Science:

• Crystallography and structural science
• Computational science
• Data science

Background:

• Machine learning (ML) is increasingly applied in scientific research.
• Crystallography and structural science generate large datasets suitable for ML analysis.
• A growing body of literature on ML in these fields is emerging.

Purpose of the Study:

• To provide an overview of a virtual collection of articles on ML in crystallography and structural science.
• To introduce fundamental artificial intelligence (AI) and ML concepts.
• To present a historical context of ML applications within IUCr journals.

Main Methods:

• Literature review of articles within the IUCr journal family (Acta Crystallographica Sections A, B, D, IUCrJ, Journal of Synchrotron Radiation).
• Conceptual explanation of key AI and ML terms.
• Historical analysis of ML's emergence in structural science publications.

Main Results:

• Identification and curation of a virtual collection focusing on ML in crystallography and structural science.
• Accessible definitions of core AI and ML terminology.
• A timeline illustrating the historical integration of ML into structural science research.

Conclusions:

• The virtual collection serves as a valuable resource for researchers entering the field.
• Understanding AI and ML fundamentals is crucial for leveraging these tools in structural science.
• The historical perspective highlights the progressive adoption and impact of ML in the discipline.