AssayRouter: Historical Utility Priors for Frozen Molecular Predictor Routing
Abstract
Laboratories often face a new molecular assay with 16-64 labels and a bank of predictors whose training data and parameters are unavailable. The practical question is which frozen outputs to include in a small local model. AssayRouter treats completed assays as pseudo-targets and labels each candidate by its post-fit utility: the reduction in held-out discovery loss when the candidate is added to the local target predictor. A shared regressor learns to predict this utility from candidate behavior on the support set, without source identity; on a new assay, one frozen ranking selects four sources and separate labels fit a convex combiner. We train only on completed ChEMBL-MT assays and evaluate 24 external regression assays across six frozen interface families. AssayRouter-C lowers strict four-call negative log-likelihood (NLL) by 0.0409 relative to Support-CV@4. Frozen candidate-label permutations confirm that candidate-utility correspondence carries the transferred information, and leave-one-interface-out training shows that the mapping generalizes to unseen predictor families. Completed assays therefore provide transferable supervision for scarce-label routing through frozen prediction interfaces.
Related
- KAT-Net: a hyperbolic routing framework for joint lysine acylation site prediction with shared sequence–structure representations
- A 209-Site DNA Methylation Predictor of Chronological Age Evaluated Under Empirical and Random Missingness.
- StrucNS reveals interaction-weighted network topology as the driving predictor of absolute stability of natural and de novo proteins
- Lipid droplet-mitochondria crosstalk in osteoarthritis: linking lipid routing to metabolic mismatch.
- Historical Areas of Hybridization in Two Striped Emerald Dragonfly Species
BibTeX
@misc{xu2026assayrouter,
title = {{AssayRouter: Historical Utility Priors for Frozen Molecular Predictor Routing}},
author = {Dong Xu and Zhangfan Yang and Jiantao Wu and Shipeng Zhang and Zexuan Zhu and Jiangqiang Li and Jun Zhang and Junkai Ji},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.37285},
eprint = {2609.37285},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.37285},
}
RIS
TY - UNPB
TI - AssayRouter: Historical Utility Priors for Frozen Molecular Predictor Routing
AU - Dong Xu
AU - Zhangfan Yang
AU - Jiantao Wu
AU - Shipeng Zhang
AU - Zexuan Zhu
AU - Jiangqiang Li
AU - Jun Zhang
AU - Junkai Ji
PY - 2026
DA - 2026/09/29
PB - arXiv
DO - 10.48550/arxiv.2609.37285
UR - https://arxiv.org/abs/2609.37285
AB - Laboratories often face a new molecular assay with 16-64 labels and a bank of predictors whose training data and parameters are unavailable. The practical question is which frozen outputs to include in a small local model. AssayRouter treats completed assays as pseudo-targets and labels each candidate by its post-fit utility: the reduction in held-out discovery loss when the candidate is added to the local target predictor. A shared regressor learns to predict this utility from candidate behavior on the support set, without source identity; on a new assay, one frozen ranking selects four sources and separate labels fit a convex combiner. We train only on completed ChEMBL-MT assays and evaluate 24 external regression assays across six frozen interface families. AssayRouter-C lowers strict four-call negative log-likelihood (NLL) by 0.0409 relative to Support-CV@4. Frozen candidate-label permutations confirm that candidate-utility correspondence carries the transferred information, and leave-one-interface-out training shows that the mapping generalizes to unseen predictor families. Completed assays therefore provide transferable supervision for scarce-label routing through frozen prediction interfaces.
ER -