CliffRank: A Dual-Branch Framework for Activity-Cliff Ranking Prediction
Abstract
Activity-cliff ranking remains difficult because local structural changes can cause large activity differences, while high-quality data that resolve the underlying mechanisms remain limited. To use available activity labels more effectively, we combine absolute-activity regression with ranking-consistency learning. CliffRank trains two parallel predictors with mean squared error, a thresholded listwise loss, and Pairwise Preference Consistency (PPC), which aligns relative ordering in the preference-probability space. On three antimicrobial peptide datasets, CliffRank with ESM2-t12 achieved the highest mean Spearman correlation of 0.5393 and mean Recall@50 of 21.4, although the leading method varied across individual datasets. On three small-molecule datasets, CliffRank with PNA, where PPC was activated after 120 epochs, achieved the highest mean Spearman correlation of 0.6890, while its mean Recall@50 of 30.4 matched that of ACANet-PNA. The PPC results also define its practical limits. Asymmetric initialization improved the MolCLR-GIN averages but did not improve every target. For PNA without pretrained weights, delayed PPC improved selected metrics, but no schedule was best for both mean Spearman correlation and mean Recall@50. Future work should evaluate more targets and antimicrobial peptide systems, develop adaptive PPC schedules, and incorporate protein or membrane context when available.
Related
- AbLWR: a context-aware listwise ranking framework for antibody-antigen binding affinity prediction via positive-unlabeled learning
- A deep learning framework for threshold-free relative gene expression ranking from histone modification signals.
- Rapid classification of β-lactamase activity through AI-based structure prediction and electric field analysis
- An integrative multi-omics modeling framework for chemotherapy response prediction through microRNA–target regulatory signatures
- KAT-Net: a hyperbolic routing framework for joint lysine acylation site prediction with shared sequence–structure representations
BibTeX
@misc{li2026cliffrank,
title = {{CliffRank: A Dual-Branch Framework for Activity-Cliff Ranking Prediction}},
author = {Kewei Li and Rongying Zhang and Peiyu Yang and Zhongjian Wang and Qiuchen Zhao and Lan Huang and Fengfeng Zhou},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.01673},
eprint = {2609.01673},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.01673},
note = {CC BY-NC-SA 4.0},
}
RIS
TY - UNPB
TI - CliffRank: A Dual-Branch Framework for Activity-Cliff Ranking Prediction
AU - Kewei Li
AU - Rongying Zhang
AU - Peiyu Yang
AU - Zhongjian Wang
AU - Qiuchen Zhao
AU - Lan Huang
AU - Fengfeng Zhou
PY - 2026
DA - 2026/09/01
PB - arXiv
DO - 10.48550/arxiv.2609.01673
UR - https://arxiv.org/abs/2609.01673
AB - Activity-cliff ranking remains difficult because local structural changes can cause large activity differences, while high-quality data that resolve the underlying mechanisms remain limited. To use available activity labels more effectively, we combine absolute-activity regression with ranking-consistency learning. CliffRank trains two parallel predictors with mean squared error, a thresholded listwise loss, and Pairwise Preference Consistency (PPC), which aligns relative ordering in the preference-probability space. On three antimicrobial peptide datasets, CliffRank with ESM2-t12 achieved the highest mean Spearman correlation of 0.5393 and mean Recall@50 of 21.4, although the leading method varied across individual datasets. On three small-molecule datasets, CliffRank with PNA, where PPC was activated after 120 epochs, achieved the highest mean Spearman correlation of 0.6890, while its mean Recall@50 of 30.4 matched that of ACANet-PNA. The PPC results also define its practical limits. Asymmetric initialization improved the MolCLR-GIN averages but did not improve every target. For PNA without pretrained weights, delayed PPC improved selected metrics, but no schedule was best for both mean Spearman correlation and mean Recall@50. Future work should evaluate more targets and antimicrobial peptide systems, develop adaptive PPC schedules, and incorporate protein or membrane context when available.
ER -