Biology subjects

Heng, P. A.

Publications and source records attributed to Heng, P. A..

6 recordsLinked to original sources

bioRxiv2026

SubcellPert: Benchmarking Subcellular Perturbation Prediction

Perturbations change both RNA abundance and intracellular organization, but gene-count predictors omit the positions needed to model spatial responses. To address this limitation, we introduce \textbf{\textit{SubcellPert}}, a benchmark for perturbation-response prediction at single-RNA-molecule resolution. It comprises 14 imaging datasets with 13.88 million cells and 3.65 billion RNA molecules. Given source observations and a directed condition change, models generate target RNA populations in normalized two-dimensional coordinates. Two independently trained tasks separate abundance from localization: \textit{Oracle} predicts coordinates given exact target counts; \textit{Joint} predicts both. Eight baselines span source resampling, classical transport, and frozen foundation-model adapters. Fixed perturbation and cellular-context holdouts evaluate abundance error, target-gene coverage, and conditional spatial error. Through within-dataset analyses, we find context-dependent gains from spatial features in condition decoding. Interpreting spatial responses requires distinguishing changes within spatial clusters from population composition, and RNA fractions from enrichment relative to compartment geometry. Through comprehensive experiments, we find that Gaussian transport achieves the lowest Oracle spatial error on eight of eleven cellular-context OOD datasets. Audits show that better centroid accuracy can accompany worse radial distributions, and that 19 of 190 Joint adapter fits retain source counts after validation. SubcellPert opens a spatial frontier for virtual cell research, laying the groundwork for predictive models that capture both the molecular composition and intracellular organization of cellular responses. The code is available at \url{https://github.com/king159/SubcellPert}.

bioinformatics↗

bioRxiv2026

NACraft: Programmatic nucleic-acid aptamer design via all-atom structure-model feedback

Protein-nucleic-acid interactions underpin diverse biological processes and provide a basis for molecular sensing, regulation and therapeutic intervention. However, the coupled dependence of aptamer function on nucleotide sequence, three-dimensional folding and target binding makes rational RNA and DNA binder design challenging. Here we present NACraft, a training-free and programmatic framework for all-atom nucleic-acid aptamer design based on backpropagation through structure-model feedback. By composing binding, sequence-similarity and anti-binding constraints, NACraft supports de novo generation, similarity-guided sampling and target-selective design within a unified optimization framework, without task-specific training or fine-tuning. Computational experiments showed that NACraft generated high-confidence candidates de novo across diverse protein targets, with further improvements achieved through similarity-guided design for both RNA and DNA complexes. Its target-selective design capability was further validated in silico, with 69.44% of paired candidates generated to favour the positive target EGFR over the off-target HER2. Under matched independent AlphaFold3 evaluation, NACraft achieved better performance than ODesign in 10 of 11 NA-12 targets and 17 of 20 protein target-length settings. Together, these results demonstrate the effectiveness and versatility of NACraft and extend structure-model hallucination toward programmatic nucleic-acid aptamer design. Codehttps://github.com/OTEAM-AI4S/NACraft

bioinformatics↗

bioRxiv2026

An Isoform-Centric, Structure-Aware Framework for Protein Function Prediction and Evaluation, Instantiated in 3DisoDeepPF

Understanding and accurately predicting function across protein isoforms has been a long-standing challenge with profound implications for both biological and translational research. However, most functional annotations and benchmarks remain tied to genes or reference protein sequences, leaving limited support for distinguishing protein isoforms from the same gene whose sequence, structure and domain composition may lead to different functions. To address this gap, we developed an isoform-centric framework for protein function and domain annotation that integrates a dense graph of sequence and structure similarity with protein features. We implemented this framework as 3DisoDeepPF, a graph-based multimodal model, and applied it to a breast cancer isoform atlas. Across canonical benchmarks and evaluations at isoform resolution, 3DisoDeepPF showed strong performance in predicting GO terms and Pfam domains and remained robust in tests with homology control. It further captured changes in Pfam domain composition among isoforms from the same gene, including reference-relative domain gain and loss. An evidence tracing module links predicted labels to supporting proteins in the graph, as illustrated by the calcium and integrin binding protein 1 (CIB1). Together, this study provides a framework informed by protein structure for function and domain annotation at isoform resolution, converting human protein isoform diversity into traceable functional evidence for cancer atlas interpretation.

cancer biology↗

bioRxiv2024

siRNADesign: A Graph Neural Network for siRNA Efficacy Prediction via Deep RNA Sequence Analysis

The clinical adoption of small interfering RNAs (siRNAs) has prompted the development of various computational strategies for siRNA design, from traditional data analysis to advanced machine learning techniques. However, previous studies have inadequately considered the full complexity of the siRNA silencing mechanism, neglecting critical elements such as siRNA positioning on mRNA, RNA base-pairing probabilities, and RNA-AGO2 interactions, thereby limiting the insight and accuracy of existing models. Here, we introduce siRNADesign, a Graph Neural Network (GNN) framework that leverages both non-empirical and empirical rules-based features of siRNA and mRNA to effectively capture the complex dynamics of gene silencing. In multiple internal datasets, siRNADesign achieves state-of-the-art performance. Significantly, siRNADesign also outperforms existing methodologies in in vitro wet lab experiments and an externally validated dataset. Additionally, we develop a new data-splitting methodology that addresses the data leakage issue, a frequently overlooked issue in previous studies, ensuring the robustness and stability of our model under various experimental settings. Through rigorous testing, siRNADesign has demonstrated remarkable predictive accuracy and robustness, making significant contributions to the field of gene silencing. Furthermore, our approach in redefining data-splitting standards aims to set new benchmarks for future research in the domain of predictive biological modeling for siRNA.

bioinformatics↗

bioRxiv2024

InstructPLM: Aligning Protein Language Models to Follow Protein Structure Instructions

Large language models are renowned for their efficacy in capturing intricate patterns, including co-evolutionary relationships, and underlying protein languages. However, current methodologies often fall short in illustrating the emergence of genomic insertions, duplications, and insertion/deletions (indels), which account for approximately 14% of human pathogenic mutations. Given that structure dictates function, mutated proteins with similar structures are more likely to persist throughout biological evolution. Motivated by this, we leverage crossmodality alignment and instruct fine-tuning techniques inspired by large language models to align a generative protein language model with protein structure instructions. Specifically, we present a method for generating variable-length and diverse proteins to explore and simulate the complex evolution of life, thereby expanding the repertoire of options for protein engineering. Our proposed protein LM-based approach, InstructPLM, demonstrates significant performance enhancements both in silico and in vitro. On native protein backbones, it achieves a perplexity of 2.68 and a sequence recovery rate of 57.51, surpassing Protein-MPNN by 39.2% and 25.1%, respectively. Furthermore, we validate the efficacy of our model by redesigning PETase and L-MDH. For PETase, all fifteen designed variable-length PETase exhibit depolymerization activity, with eleven surpassing the activity levels of the wild type. Regarding L-MDH, an enzyme lacking an experimentally determined structure, InstructPLM is able to design functional enzymes with an AF2-predicted structure. Code and model weights of InstructPLM are publicly available*.

bioinformatics↗

bioRxiv2023

Protein Sequence Design by Entropy-based Iterative Refinement

Inverse Protein Folding (IPF) is an important task of protein design, which aims to design sequences compatible with a given backbone structure. Despite the prosperous development of algorithms for this task, existing methods tend to leverage limited and noisy residue environment when generating sequences. In this paper, we develop an iterative sequence refinement pipeline, which can refine the sequence generated by existing sequence design models. It selects and retains reliable predictions based on the models confidence in predicted distributions, and decodes the residue type based on a partially visible environment. The proposed scheme can consistently improve the performance of a number of IPF models on several sequence design benchmarks, and increase sequence recovery of the SOTA model by up to 10%. We finally show that the proposed model can be applied to redesign Transposon-associated transposase B. 8 variants exhibit improved gene editing activity among the 20 variants we proposed. Our code and a demo of the refinement pipeline are provided in the online colab.

bioinformatics↗