siProGenA: Generative siRNA Candidate Construction via Position Proposal and Guide Generation
- 1School of Engineering, Westlake University
Abstract
Small interfering RNAs (siRNAs) are short guide RNAs that recruit the RNA-induced silencing complex (RISC) to complementary target sites on messenger RNAs (mRNAs), triggering Ago2-mediated cleavage and gene silencing. siRNA design requires compact candidate sets that cover a target while preserving efficacy, specificity, and practical sequence constraints. Existing pipelines usually enumerate candidate windows, assign a canonical guide to each window, and then rank preconstructed siRNA--mRNA pairs. This has produced strong pairwise efficacy predictors, but leaves a candidate-construction gap: candidate positions and guide sequences are fixed before the model begins to rank them. We address this gap by decomposing siRNA candidate construction into two generative decisions: where to place candidates within an mRNA segment, and what constrained guide variants to consider at a candidate position. We instantiate this framework as siProGenA, using a Discrete Denoising Diffusion Probabilistic Model (D3PM) for mRNA-conditioned position proposal and a Bayesian Flow Network (BFN) for temperature-controlled guide generation. On 62 positive test segments, the diversity-aware final library reaches Hit@1 = 0.790 and Hit@5 = 0.903. In a measured-site controlled Stage~2 evaluation, seed- and cleavage-preserving variants outscore the canonical complement for 89.8% of measured sites, with supporting gains across additional computational scorers, random-mismatch controls, and biophysical diagnostics. Together, the results support a modular proposal--generation view of siRNA candidate construction for prioritizing compact candidate sets.
Related
- Human position sense based on memories is not proprioception
- Delocalized DNA holes imprint mutation position in mismatch-repair-deficient cancers
- Ten simple rules for preparing a job application for an academic position
- The European Anthropological Association (EAA) position statement on anthropometry: education, competence, and scientific authority
- Tonic neuron position and network topology determine synchronization dynamics
BibTeX
@misc{ma2026siprogena,
title = {{siProGenA: Generative siRNA Candidate Construction via Position Proposal and Guide Generation}},
author = {Z. Ma and J. Zhou and R. Wang and Z. Deng and Z. Wu and Y. Zheng},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.31.748303},
url = {https://www.biorxiv.org/content/10.64898/2026.08.31.748303v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - siProGenA: Generative siRNA Candidate Construction via Position Proposal and Guide Generation
AU - Z. Ma
AU - J. Zhou
AU - R. Wang
AU - Z. Deng
AU - Z. Wu
AU - Y. Zheng
PY - 2026
DA - 2026/08/31
PB - bioRxiv
DO - 10.64898/2026.08.31.748303
UR - https://www.biorxiv.org/content/10.64898/2026.08.31.748303v1
AB - Small interfering RNAs (siRNAs) are short guide RNAs that recruit the RNA-induced silencing complex (RISC) to complementary target sites on messenger RNAs (mRNAs), triggering Ago2-mediated cleavage and gene silencing. siRNA design requires compact candidate sets that cover a target while preserving efficacy, specificity, and practical sequence constraints. Existing pipelines usually enumerate candidate windows, assign a canonical guide to each window, and then rank preconstructed siRNA--mRNA pairs. This has produced strong pairwise efficacy predictors, but leaves a candidate-construction gap: candidate positions and guide sequences are fixed before the model begins to rank them. We address this gap by decomposing siRNA candidate construction into two generative decisions: where to place candidates within an mRNA segment, and what constrained guide variants to consider at a candidate position. We instantiate this framework as siProGenA, using a Discrete Denoising Diffusion Probabilistic Model (D3PM) for mRNA-conditioned position proposal and a Bayesian Flow Network (BFN) for temperature-controlled guide generation. On 62 positive test segments, the diversity-aware final library reaches Hit@1 = 0.790 and Hit@5 = 0.903. In a measured-site controlled Stage~2 evaluation, seed- and cleavage-preserving variants outscore the canonical complement for 89.8% of measured sites, with supporting gains across additional computational scorers, random-mismatch controls, and biophysical diagnostics. Together, the results support a modular proposal--generation view of siRNA candidate construction for prioritizing compact candidate sets.
ER -