PAIR-Former: Budgeted Relational MIL for miRNA Target Prediction

2026-01-31

AI Summary

This work addresses the challenge of miRNA–mRNA target prediction, where an enormous number of candidate binding sites coexists with scarce pairwise labels and limited computational resources. The authors propose the first Budgeted Relational Multi-Instance Learning (BR-MIL) framework, which operates under a strict computational budget by efficiently encoding and modeling relationships among at most K candidate sites. The approach employs a two-stage pipeline: a low-cost scan over the entire candidate pool followed by a diversity-based selection of K sites on CPU, whose interdependencies are captured via a permutation-invariant Set Transformer aggregator. Evaluated on the miRAW dataset under a practical budget of K=64, the method significantly outperforms strong baselines while achieving high accuracy, controllable computational overhead, and theoretical guarantees, thereby demonstrating its effectiveness and scalability.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Search and Optimization: Metareasoning and Metaheuristics
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Functional miRNA--mRNA targeting is a large-bag prediction problem: each transcript yields a heavy-tailed pool of candidate target sites (CTSs), yet only a pair-level label is observed. We formalize this regime as \emph{Budgeted Relational Multi-Instance Learning (BR-MIL)}, where at most $K$ instances per bag may receive expensive encoding and relational processing under a hard compute budget. We propose \textbf{PAIR-Former} (Pool-Aware Instance-Relational Transformer), a BR-MIL pipeline that performs a cheap full-pool scan, selects up to $K$ diverse CTSs on CPU, and applies a permutation-invariant Set Transformer aggregator on the selected tokens. On miRAW, PAIR-Former outperforms strong pooling baselines at a practical operating budget ($K^\star{=}64$) while providing a controllable accuracy--compute trade-off as $K$ varies. We further provide theory linking budgeted selection to (i) approximation error decreasing with $K$ and (ii) generalization terms governed by $K$ in the expensive relational component.

Problem

• miRNA target prediction
• Budgeted Relational Multi-Instance Learning
• candidate target sites
• compute budget
• large-bag prediction

Innovation

• Budgeted Relational MIL
• PAIR-Former
• miRNA target prediction
• Set Transformer
• compute-efficient selection