PUFFIN: Protein Unit Discovery with Functional Supervision

2026-04-16

AI Summary

Existing methods struggle to identify biologically meaningful mesoscale functional units from protein structures while effectively integrating functional information. This work proposes a novel approach that, for the first time, incorporates functional label supervision into the structure-based graph partitioning process without relying on predefined annotations. By leveraging graph neural networks and a structure-aware pooling mechanism, the method jointly learns to partition residue-level structural graphs and extract functional signals, thereby automatically discovering structurally coherent, multi-residue protein functional units. The identified units exhibit strong statistical associations with molecular functions and demonstrate high concordance with manually curated InterPro annotations, enabling interpretable modeling of structure–function relationships.

Technology Category

• Machine Learning: Structured Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Web Mining and Content Analysis: Bridging structured and unstructured data

Abstract

Proteins carry out biological functions through the coordinated action of groups of residues organized into structural arrangements. These arrangements, which we refer to as protein units, exist at an intermediate scale, being larger than individual residues yet smaller than entire proteins. A deeper understanding of protein function can be achieved by identifying these units and their associations with function. However, existing approaches either focus on residue-level signals, rely on curated annotations, or segment protein structures without incorporating functional information, thereby limiting interpretable analysis of structure-function relationships. We introduce PUFFIN, a data-driven framework for discovering protein units by jointly learning structural partitioning and functional supervision. PUFFIN represents proteins as residue-level structure graphs and applies a graph neural network with a structure-aware pooling mechanism that partitions each protein into multi-residue units, with functional supervision that shapes the partition. We show that the learned units are structurally coherent, exhibit organized associations with molecular function, and show meaningful correspondence with curated InterPro annotations. Together, these results demonstrate that PUFFIN provides an interpretable framework for analyzing structure-function relationships using learned protein units and their statistical function associations. We made our source code available at https://github.com/boun-tabi-lifelu/puffin.

Problem

• protein units
• structure-function relationship
• functional supervision
• protein structure partitioning
• interpretable analysis

Innovation

• protein units
• graph neural network
• structure-function relationship
• functional supervision
• structure-aware pooling