MIRCID: Inferred Hub-miRNAs Drive Cross-Task Improvements in Drug Mechanistic Modeling
- 1School of Medicine, The Chinese University of Hong Kong, Shenzhen, Longgang District, Shenzhen, Guangdong 518172, China; Warshel Institute for Computational Bio
Abstract
Drug mechanism-of-action (MoA) modeling commonly relies on perturbational transcriptomes, but matched microRNA (miRNA) measurements are often unavailable. Inferred regulatory features offer a scalable way to reuse these data. Here, we present MIRCID, a framework comparing gene expression with inferred transcription factor (TF) activity and miRNA expression across pathway classification and similarity-based MoA retrieval. HubmiRNet infers 414 pan-cancer hub miRNAs (HubmiRs) from 977 L1000 landmark genes, achieving a Pearson correlation coefficient of 87.72\%; its 1,298-output variant also outperformed SiCmiR on the full-miRNA task (71.21\% versus 67.30\%). In the evaluated comparisons, miRNA augmentation provided more consistent gains than TF activity. Generic embedding controls showed model-dependent utility, while complementarity analyses identified a distinct, partially linearly recoverable representation that retained gene-derived structure. Illustrative rescue cases linked improved classification to biologically plausible miRNA patterns in samples with weak transcriptional signatures. These findings support inferred HubmiRs as a biologically informed recoding of transcriptomic data for perturbational drug modeling, while leaving recovery of measured perturbational miRNA responses to further validation.
Related
- Telomere dynamics parameters inferred from mathematical modeling of telomere length distribution
- Mechanistic modeling of bacterial translation initiation across growth conditions
- FEED, FLEE, OR FACT-FIND: MODELING LATENT COGNITIVE-MECHANISTIC RESPONSES TO HETEROSPECIFIC ALARM CALLS FOR FORAGING CARDINALS
- Unraveling Trimethoprim-Sulfonamide synergy against Escherichia coli through semi-mechanistic PK/PD modeling
- A mechanistic modeling framework to interpret ACTH stimulation tests across HPA axis adaptation states and glucocorticoid feedback dynamics
BibTeX
@misc{cao2026mircid,
title = {{MIRCID: Inferred Hub-miRNAs Drive Cross-Task Improvements in Drug Mechanistic Modeling}},
author = {X. Cao and Y. Chen and J. Xu and Z. Zhang and X. Cheng and S. Wang and Y. Zhang and X. Cai and S. Cui and Z. Zhu and X. Ji and H.-Y. Huang and Y.-C.-D. Lin and H.-D. Huang},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.17.752312},
url = {https://www.biorxiv.org/content/10.64898/2026.09.17.752312v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - MIRCID: Inferred Hub-miRNAs Drive Cross-Task Improvements in Drug Mechanistic Modeling
AU - X. Cao
AU - Y. Chen
AU - J. Xu
AU - Z. Zhang
AU - X. Cheng
AU - S. Wang
AU - Y. Zhang
AU - X. Cai
AU - S. Cui
AU - Z. Zhu
AU - X. Ji
AU - H.-Y. Huang
AU - Y.-C.-D. Lin
AU - H.-D. Huang
PY - 2026
DA - 2026/09/17
PB - bioRxiv
DO - 10.64898/2026.09.17.752312
UR - https://www.biorxiv.org/content/10.64898/2026.09.17.752312v1
AB - Drug mechanism-of-action (MoA) modeling commonly relies on perturbational transcriptomes, but matched microRNA (miRNA) measurements are often unavailable. Inferred regulatory features offer a scalable way to reuse these data. Here, we present MIRCID, a framework comparing gene expression with inferred transcription factor (TF) activity and miRNA expression across pathway classification and similarity-based MoA retrieval. HubmiRNet infers 414 pan-cancer hub miRNAs (HubmiRs) from 977 L1000 landmark genes, achieving a Pearson correlation coefficient of 87.72\%; its 1,298-output variant also outperformed SiCmiR on the full-miRNA task (71.21\% versus 67.30\%). In the evaluated comparisons, miRNA augmentation provided more consistent gains than TF activity. Generic embedding controls showed model-dependent utility, while complementarity analyses identified a distinct, partially linearly recoverable representation that retained gene-derived structure. Illustrative rescue cases linked improved classification to biologically plausible miRNA patterns in samples with weak transcriptional signatures. These findings support inferred HubmiRs as a biologically informed recoding of transcriptomic data for perturbational drug modeling, while leaving recovery of measured perturbational miRNA responses to further validation.
ER -