QUAM-AFM: A Free Database for Molecular Identification by Atomic Force Microscopy

Area of Science:

• Computational Chemistry
• Materials Science
• Data Science

Background:

• Atomic Force Microscopy (AFM) is crucial for molecular imaging.
• Accurate chemical identification from AFM images remains a challenge.
• Large-scale, diverse datasets are needed to train advanced analytical models.

Purpose of the Study:

• Introduce the Quasar Science Resources-Autonomous University of Madrid atomic force microscopy image data set (QUAM-AFM).
• Provide a comprehensive resource for advancing chemical identification in AFM experiments.
• Facilitate the application of deep learning techniques to AFM data analysis.

Main Methods:

• Generation of 165 million simulated AFM images from 685,513 organic molecules.
• Creation of 3D image stacks with varying operational parameters and tip-sample distances.
• Development of a graphical user interface for data retrieval and searching.

Main Results:

• The QUAM-AFM dataset is the largest collection of simulated AFM images to date.
• Each molecule is represented by 24 3D image stacks at 256x256 resolution.
• Associated data includes chemical structures, IUPAC names, and atomic coordinates.

Conclusions:

• The QUAM-AFM dataset is a valuable resource for deep learning-based chemical identification in AFM.
• The dataset's scale and diversity support robust model training.
• The provided interface enhances accessibility for researchers.