Biology subjects

Bu, D.

Publications and source records attributed to Bu, D..

At least 19 recordsLinked to original sources

bioRxiv2026

ComplexDesign: sequence-hallucination design of protein binders bridging multiple proteins

MotivationDesigning multichain protein complexes requires coordinating the folding of component proteins with the formation of their interfaces. The existing methods, however, remain limited in their ability to satisfy these requirements simultaneously, especially for trimeric and tetrameric complexes. As an important practical scenario, designing a binder that bridges two target proteins into a ternary complex requires flexibility in the relative arrangement of the two targets, adding an additional challenge to existing design methods. ResultsWe present ComplexDesign, a hallucination-based approach for multichain protein design. ComplexDesign performs structure-prediction-guided sequence optimization to simultaneously fold each protein chain and form inter-chain interactions that bind them together. To provide the flexibility required to appropriately arrange these target proteins, ComplexDesign introduces a specialized masking mechanism that enables exploration of possible relative arrangements rather than being limited to the predefined ones. Across a comprehensive set of benchmarks with various chain lengths, ComplexDesign outperformed existing methods in the unconditional design of dimers, trimers, and tetramers, achieving a high design success rate exceeding 50%, supporting its capability for multichain complex design. Furthermore, in the case of multi-target binder design, ComplexDesign produced high-confidence, self-consistent ternary complexes for 8 out of 10 target pairs. These results establish ComplexDesign as an effective tool for multichain protein design, with particular utility for designing binders that bridge two target proteins. Availability and implementationThe source code of ComplexDesign will be made publicly available upon publication.

bioinformatics↗

bioRxiv2026

Objective curriculum-guided design of multi-property proteins

Designing functional proteins that simultaneously possess multiple bio-chemical properties remains a significant challenge, as key protein properties, such as solubility, stability, binding affinity, and chemical resistance, are often interdependent or even conflicting. Current approaches typically attempt to jointly optimize multiple functional objectives in one shot, followed by extensive screening to identify rare feasible designs. Here, we introduce OCDesign, an objective curriculum-guided framework for multi-property protein design. OCDesign is based on the objective curriculum principle, i.e., the order in which objectives are introduced can shape the accessibility of functional solutions. In each design round of OCDesign, candidate sequences are generated in silico, assessed across multiple properties, selected based on Pareto-optimal trade-offs, and experimentally validated, with each experimental stage testing the role of the newly introduced objective within the curriculum. Using antibody-binding protein A as a model system, we show that one-shot optimization fails to yield functional designs, whereas a staged curriculum--progressing from solubility and structural consistency to binding affinity, and then to alkaline resistance--enables the design of proteins possessing multiple desired properties through substantially fewer wet-lab experiments. These results establish OCDesign as a practical computational-experimental strategy for organizing and integrating multiple objectives in protein design, and suggest that objective ordering is a key determinant of accessibility in high-dimensional design spaces.

bioengineering↗

bioRxiv2026

Sampling protein structural token space enables accurate prediction of multiple conformations

Protein function is fundamentally mediated by ensembles of distinct metastable states. However, existing methods, such as AlphaFold 3, typically exhibit a bias toward predicting a single dominant state, failing to capture alternative conformations or provide robust metrics for identifying high-quality multi-state conformations. Here, we present MultiStateFold (MSFold), a framework that integrates Parallel Tempering into the discrete structure token space of the ESM3 protein language model. By conceptualizing the model's latent space as an implicit energy landscape, MSFold enables global exploration and barrier crossing, thereby overcoming the local sampling limitations inherent in base generative models. Across a benchmark of 313 multi-conformation pairs, MSFold sets a new performance standard: it achieves the highest success rate in modeling native states and substantially outperforms leading methods, including AlphaFold 3, on challenging alternative conformations, while maintaining competitive accuracy for primary structures. Furthermore, we propose Sequence Log-Likelihood (SLL), a novel confidence metric derived from sequence-structure consistency. Our results demonstrate that SLL offers a modest improvement over standard metrics such as pTM and pLDDT. This work establishes a new paradigm for conformational sampling, bridging classical statistical physics with protein language models.

bioinformatics↗

bioRxiv2026

Predicting TCR-pMHC Binding by Reinforcement Learning

The binding between T cell receptors (TCRs) and peptide-major histocompat-ibility complexes (pMHCs) is fundamental to the immune systems ability to recognize and eliminate pathogens. Accurate prediction of TCR-pMHC inter-actions holds significant promise for advancing cancer immunotherapy, vaccine design, and autoimmune disease research. However, existing approaches often treat the sequences, structures, and functions of TCRs, peptides, and MHC molecules in isolation, neglecting their interdependencies and hence limiting the prediction accuracy. In this study, we present ProTCR, a novel approach that integrates sequence, structural, and functional information within a reinforce-ment learning framework, offering a new paradigm for predicting TCR-pMHC binding. The reinforcement learning optimization enables ProTCR to generate TCR-pMHC sequences with enhanced binding propensity, thereby improving prediction accuracy. On benchmark datasets such as IEDB and VDJdb, ProTCR achieves an AUROC of 0.75, outperforming state-of-the-art methods by 32.7%, while offering interpretable insights into the structural and sequence determi-nants of binding. We further validate ProTCR using TCRs and neoantigens derived from a cervical cancer patient via proteogenomic profiling. Our analysis reveals a strong correlation between T cell clonal expansion and ProTCR-predicted TCR-peptide binding scores, supporting the biological relevance of the model. Additionally, ProTCR demonstrates robust performance in predict-ing SARS-CoV-2 TCR-pMHC complexes and generating MHC-specific peptides with potential applications in peptide-based immunotherapies. Collectively, these findings establish ProTCR as a powerful and interpretable tool for TCR-pMHC binding prediction, with broad utility across immunology research and translational applications.

bioinformatics↗

bioRxiv2025

ProAffinity++

Proteins are essential biological macromolecules that play a crucial role in living organisms. Protein-protein interactions, which govern various biological processes such as signal transduction, cell metabolism, and cell growth, are key aspect of protein function. The strength of these interactions, characterized by protein-protein binding affinity({Delta}G), is a critical thermodynamic property of protein complexes. Accurate prediction of protein-protein binding affinity is meaningful for understanding the mechanisms of biological systems and improving the speed of binding affinity determination. However, current methods are limited in their accuracy and are primarily designed for binary complexes. To address these limitations, we propose ProAffinity++, an end-to-end algorithm that predicts binding affinities using a combined representation of sequence and structure. The core idea of ProAffinity++ is to model the binding regions of protein complexes and utilize graph neural networks to capture the local microenvironment of residues. Our experimental evaluations demonstrate that ProAffinity++ outperforms state-of-the-art methods in predicting protein-protein binding affinities. Moreover, it exhibits remarkable performance in challenging scenarios, such as antigen-antibody interactions and missense mutation problems, where existing methods have limitations.

bioinformatics↗

bioRxiv2025

Accurate and efficient phylogenetic inference through end-to-end deep learning

Accurate phylogenetic inference is crucial for understanding evolutionary relationships among species. Deep learning technique has been introduced for phylogenetic inference; however, the existing deep learning-based approaches either suffer from limited accuracy as they split inference into several disjoint stages, or exhibit low efficiency and hardly apply to the cases with over 20 species. Here we present an accurate and efficient approach to phylogenetic inference. Our approach, called NeuralNJ, employs an end-to-end framework that directly constructs phylogenetic trees from the input taxa, thus effectively avoiding the inaccuracy incurred by the split inference stages. The key innovation of NeuralNJ lies in its learnable neighbor joining mechanism, which iteratively joins neighbors guided by learned priority scores and thereby achieves accurate tree reconstruction. The inference accuracy is further enhanced through incorporating reinforcement learning-based tree search. Using both simulated and empirical data as representatives, we demonstrate that NeuralNJ can effectively infer phylogenetic tree with improved computational efficiency and reconstruction accuracy. The study paves the way to accurate and efficient phylogenetic inference for hundreds of taxa in complex evolutionary scenarios.

bioinformatics↗

bioRxiv2025

DNA Data Storage Architecture via Ligation of Dynamic DNA Bytes

The explosive growth of digital data is overwhelming conventional storage media, creating an urgent need for more efficient solutions. DNA offers immense potential for digital data storage, yet most systems remain static and archival. Here, we present a modular DNA storage architecture based on dynamic DNA bytes (DynaBytes)-- pre-fabricated DNA segments that can be ligated into reconfigurable information units. Utilizing core, functional and control DynaBytes, we stored 210,776 bits (26,347 bytes) of digital information organized within a file-system, and demonstrated CRUD (Create-Read-Update-Delete)-like operations, hierarchical access and nanopore-based realtime retrieval. Robust data recovery was achieved under [~]100x error-prone sequencing using streamlined error correction and fuzzy decoding. By relying on in vitro ligation of standardized components, DynaBytes reduces cost, scales efficiently, and allows interactive, rewritable storage. These features advance DNA storage beyond passive archiving toward a reconfigurable framework, opening new possibilities for dynamic, practical and large-scale DNA-based data systems.

synthetic biology↗

bioRxiv2024

Accurate structure prediction of immune proteins using parameter-efficient transfer learning

Accurate prediction of immune protein structures is crucial for understanding the immune system and advancing immunotherapy development. While deep learning methods have significantly advanced protein structure prediction by extracting evolutionary constraints from homologous sequences of a target protein, they struggle with immune proteins due to the limited number of known structures and the lack of homologous sequences in hypervariable regions. To address this challenge, we propose ImmuneFold, a transfer learning approach that fine-tunes ESMFold specifically for immune proteins. We leverage low-rank adaption (LoRA), a parameter-efficient fine-tuning technique that requires considerably less memory and substantially fewer parameters. Evaluations on various immune proteins, including T-cell receptors, antibodies, and nanobodies, demonstrate that ImmuneFold outperforms existing methods in prediction accuracy. Furthermore, we apply ImmuneFold to develop a zero-shot protocol for TCR-epitope binding prediction. Unlike previous supervised methods suffering from severe overfitting due to limited experimental binding data, our approach first predicts TCR-epitope structure using ImmuneFold and then directly estimates the binding affinity by calculating Rosseta energy. Evaluations on experimental binding datasets suggest that our method is robust and accurate in predicting TCR-epitope binding. In summary, ImmuneFold demonstrates accurate predictions of immune protein structures and TCR-epitope binding, highlighting its potential to advance the development of immunotherapies.

bioinformatics↗

bioRxiv2024

Cost-effective DNA storage with DNA movable type

In the face of exponential data growth, DNA-based storage offers a promising solution for preserving big-data. However, most existing DNA storage methods, akin to traditional block printing, require costly chemical synthesis for each individual data file, adopting a sequential, one-time-use synthesis approach. To overcome these limitations, we introduce a novel, cost-effective "DNA-Movable-Type Storage" system, inspired by movable type printing. This system utilizes pre-fabricated DNA movable types-short, double-stranded DNA oligonucleotides encoding specific payload, address, and checksum data. These DNA-MTs are enzymatically ligated/assembled into cohesive sequences, termed "DNA movable type blocks", streamlining the assembly process with the automated BISHENG-1 DNA-MT inkjet printer. Using BISHENG-1, we successfully printed, assembled, stored and accurately retrieved 43.7 KB of data files in diverse formats (text, image, audio, and video) in vitro and in vivo, using only 350 DNA-MTs. Notably, each DNA-MT, synthesized once (2 OD), can be used up to 10,000 times, reducing costs to 122 $/MB-outperforming existing DNA storage methods. This innovation circumvents the need to synthesize entire DNA sequences encoding files from scratch, offering significant cost and efficiency advantages. Furthermore, it has considerable untapped potential to advance a robust DNA storage system, better meeting the extensive data storage demands of the big-data era.

bioinformatics↗

bioRxiv2024

Language models enable zero-shot prediction of RNA secondary structures including pseudoknots

Current deep learning-based models for predicting RNA secondary structures face challenges in achieving high generalization ability. At the same time, a vast repository of unlabeled non-coding RNA (ncRNA) sequences remains untapped for structure prediction tasks. To address this challenge, we trained RNA-km, a foundation language model that enables zero-shot prediction of RNA secondary structures including pseudoknots. For the end, we incorporated specific modifications into the language model training process, including k-mer masking strategy and relative positional encoding. RNA-km are trained on 23 million ncRNA sequences in a self-supervised manner, gaining the advantages of high generalization ability. For a target RNA sequence, we make a zero-shot secondary structure prediction with the attention maps provided by RNA-km and a specified minimum-cost flow algorithm. Our results on popular benchmark datasets demonstrate that RNA-km exhibits high generalization abilities, excelling in zero-shot predictions for RNA secondary structures. In addition, the attention maps provided by the model capture intricate structural relationships, as evidenced by accurate pseudoknot predictions and precise identification of longdistance base pairs. We anticipate that RNA-km enhances the predictive capacity and robustness of existing models, thereby improving their ability to accurately predict structures for novel RNA sequences.

bioinformatics↗

bioRxiv2023

GeneCompass: Deciphering Universal Gene Regulatory Mechanisms with Knowledge-Informed Cross-Species Foundation Model

Deciphering the universal gene regulatory mechanisms in diverse organisms holds great potential to advance our knowledge of fundamental life process and facilitate research on clinical applications. However, the traditional research paradigm primarily focuses on individual model organisms, resulting in limited collection and integration of complex features on various cell types across species. Recent breakthroughs in single-cell sequencing and advancements in deep learning techniques present an unprecedented opportunity to tackle this challenge. In this study, we developed GeneCompass, the first knowledge-informed, cross-species foundation model pre-trained on an extensive dataset of over 120 million single-cell transcriptomes from human and mouse. During pre-training, GeneCompass effectively integrates four types of biological prior knowledge to enhance the understanding of gene regulatory mechanisms in a self-supervised manner. Fine-tuning towards multiple downstream tasks, GeneCompass outperforms competing state-of-the-art models in multiple tasks on single species and unlocks new realms of cross-species biological investigation. Overall, GeneCompass marks a milestone in advancing knowledge of universal gene regulatory mechanisms and accelerating the discovery of key cell fate regulators and candidate targets for drug development.

bioinformatics↗

bioRxiv2023

YiNet: An Integrated Traditional Chinese and Western Medicine Platform for Viral Infectious Diseases

Viral infectious diseases (VIDs) impose a heavy burden on global public health. Traditional Chinese medicine (TCM) has previously and is currently contributing to the prevention and treatment of infectious diseases. Omics and information technology enable the precise identification of virus characteristics, virus-host interactions and TCM mechanisms. In this study, we constructed the YiNet platform to better integrate the novel techniques and historical experience of TCM in infectious diseases. YiNet comprises three modules: knowledge base, database and toolkit. The YiNet knowledge base involves 43 VIDs, thereby systematically integrating the knowledge regarding viruses, host symptoms and TCM and Western medicine (commonly used chemical drugs, 6,899 herbs and 2,481 formulas). The YiNet database module includes multiple databases, comprising 57,340 genome sequences of 45,791 viral strains and 5,726 multi-omics datasets such as RNA-seq, ChIP-seq and ATAC-seq from different tissues and cell models. It also integrates 1,105 real-world TCM clinical cases. We adopted visual analysis tool to investigate pathogen-host-herb relationships. To explore pharmacological mechanisms for the core herbs, we added formula data mining and network pharmacology analysis pipelines and visualisation tools. YiNet can facilitate the mechanistic study of TCM and drug development for VIDs. The YiNet platform is publicly available at http://yinet.gene.ac/.

bioinformatics↗

bioRxiv2023

Highly accurate and robust protein sequence design with CarbonDesign

Protein sequence design, the inverse problem of protein structure prediction, plays a crucial role in protein engineering. Although recent deep learning-based methods have shown promising advancements, achieving accurate and robust protein sequence design remains an ongoing challenge. Here, we present CarbonDesign, a new approach that draws inspiration from successful ingredients of AlphaFold for protein structure prediction and makes significant and novel developments tailored specifically for protein sequence design. At its core, CarbonDesign explores Inverseformer, a novel network architecture adapted from AlphaFolds Evoformer, to learn representations from backbone structures and an amortized Markov Random Fields model for sequence decoding. Moreover, we incorporate other essential AlphaFold concepts into CarbonDesign: an end-to-end network recycling technique to leverage evolutionary constraints in protein language models and a multi-task learning technique to generate side chain structures corresponding to the designed sequences. Through rigorous evaluations on independent testing data sets, including the CAMEO and recent CASP15 data sets, as well as the predicted structures from AlphaFold, we show that CarbonDesign outperforms other published methods, achieving high accuracy in sequence generation. Moreover, it exhibits superior performance on de novo backbone structures obtained from recent diffusion generative models such as RFdiffusion and FrameDiff, highlighting its potential for enhancing de novo protein design. Notably, CarbonDesign also supports zero-shot prediction of the functional effects of sequence variants, indicating its potential application in directed evolution-based design. In summary, our results illustrate CarbonDesigns accurate and robust performance in protein sequence design, making it a promising tool for applications in bioengineering.

bioinformatics↗

bioRxiv2023

Collagen remodeling leads to inflammation-free expansion of periprostatic adipose tissue and promotes prostate cancer progression

One of the most striking features of the adipose depot surrounding the prostate (periprostatic adipose tissue, PPAT) is that its accumulation is independent of body mass index. Its volume varies considerably between individual with some patients exhibiting abundant PPATs that have been correlated to occurrence of aggressive prostate cancer (PCa). However, abundant PPAT are not defined at biological levels. We used a new statistical approach to define abundant PPAT by normalizing PPAT volume to prostate volume in a cohort of 351 patients with a linear regression model. Applying this definition, we confirmed the link between abundant PPAT and PCa aggressiveness, therefore validating our approach. At biological levels, we showed that abundant PPAT exhibited extensive extracellular matrix remodeling, notably of the collagen network, decreasing the mechanical constraints in hypertrophic adipocytes leading to an inflammation free-expansion. Degradation of the most abundant collagen in AT, collagen VI was associated with increased production of endotrophin, a signaling peptide derived from AT, that was also elevated in the urine of patients with abundant PPAT confirming the clinical relevance of our results. These results highlight a unique mechanism of expansion of an adipose depot and open new mechanistic avenues to explain its role in prostate-related disorders.

cancer biology↗

bioRxiv2022

Accurate prediction of RNA secondary structure including pseudoknots through solving minimum-cost flow with learned potentials

Pseudoknots are key structure motifs of RNA and pseudoknotted RNAs play important roles in a variety of biological processes. Here, we present KnotFold, an accurate approach to the prediction of RNA secondary structure including pseudoknots. The key elements of Knot-Fold include a learned potential function and a minimum-cost flow algorithm to find the secondary structure with the lowest potential. KnotFold learns the potential from the RNAs with known structures using a self-attention-based neural network, thus avoiding the inaccuracy of hand-crafted energy functions. The specially-designed minimum-cost flow algorithm used by KnotFold considers all possible combinations of base pairs and selects from them the optimal combination. The algorithm breaks the restriction of nested base pairs required by the widely-used dynamic programming algorithms, thus facilitating the identification of pseudoknots. Using a total of 1605 RNAs as representatives, we demonstrate the successful application of KnotFold in predicting RNA secondary structures including pseudoknots with accuracy significantly higher than the state-of-the-art approaches. We anticipate that KnotFold, with its superior accuracy, will greatly facilitate the understanding of RNA structures and functionalities.

bioinformatics↗

bioRxiv2022

Accurate and efficient protein sequence designthrough learning concise local environment of residues

Protein sequence design has been widely applied in rational protein engineering and increasing the design accuracy and efficiency is highly desired. Here we present ProDESIGN-LE, an accurate and efficient design approach, which adopts a concise but informative representation of residues local environment and trains a transformer to select an appropriate residue at a position from its local environment. ProDESIGN-LE iteratively applies the transformer on the positions in the target structure, eventually acquiring a designed sequence with all residues fitting well with their local environments. ProDESIGN-LE designed sequences for 68 naturally occurring and 129 hallucinated proteins within 20 seconds per protein on average, and the predicted structures from the designed sequences perfectly resemble the target structures with state-of-the-art average TM-score exceeding 0.80. We further experimentally validated ProDESIGN-LE by designing five sequences for an enzyme, chloramphenicol O-acetyltransferase type III (CAT III), and recombinantly expressing the proteins in E. coli. Of these proteins, three exhibited excellent solubility, and one yielded monomeric species with circular dichroism spectra consistent with the natural CAT III protein.

bioinformatics↗

bioRxiv2022

Combined transcriptome and metabolome analysis of a new species of microalgae from Tibetan plateau and its response to sewage treatment

Microalgae are pivotal in maintaining water quality in the lakes and rivers of Qinghai-Tibet plateau. The optimum sewage treatment conditions for Desmodesmus sp. are, temperature: 20-25{degrees}C, light intensity: 3000-8000 lx, and pH: 7.0-7.5, identified based on orthogonal experiments. The maximum removal rate of total nitrogen, total phosphorus, and chemical oxygen demand was more than 95% in the actual sewage treatment. The sewage treatment capacity of Desmodesmus sp. from plateau is higher than that from plains under the same treatment conditions. To identify the differentially expressed genes and metabolites in Desmodesmus sp. in response to sewage treatment, a combination of metabolomics and transcriptomics was employed to the microalgae with and without sewage treatment. The results showed that the oxidative phosphorylation, photosynthesis, and propanoate metabolism pathways were the most significantly enriched pathways in sewage treatment. Further, the metabolism of adenosine diphosphate, 2-oxobutanoate, and succinate were significantly upregulated, downregulated, and both upregulated and downregulated, respectively, as shown by the combined transcriptome and metabolome analysis. Additionally, we found that sewage treatment could also induce numerous changes in the primary metabolism, such as carbohydrate, fatty acid biosynthesis, and amino acid metabolism when compared with control. Overall, our results should improve fundamental knowledge of molecular responses to Desmodesmus sp. in sewage treatment and contribute to the design of strategies in microalgae response to sewage treatment.

ecology↗

bioRxiv2020

ProALIGN: Directly learning alignments for protein structure prediction via exploiting context-specific alignment motifs

Template-based modeling (TBM), including homology modeling and protein threading, is one of the most reliable techniques for protein structure prediction. It predicts protein structure by building an alignment between the query sequence under prediction and the templates with solved structures. However, it is still very challenging to build the optimal sequence-template alignment, especially when only distantly-related templates are available. Here we report a novel deep learning approach ProALIGN that can predict much more accurate sequence-template alignment. Like protein sequences consisting of sequence motifs, protein alignments are also composed of frequently-occurring alignment motifs with characteristic patterns. Alignment motifs are context-specific as their characteristic patterns are tightly related to sequence contexts of the aligned regions. Inspired by this observation, we represent a protein alignment as a binary matrix (in which 1 denotes an aligned residue pair) and then use a deep convolutional neural network to predict the optimal alignment from the query protein and its template. The trained neural network implicitly but effectively encodes an alignment scoring function, which reduces inaccuracies in the handcrafted scoring functions widely used by the current threading approaches. For a query protein and a template, we apply the neural network to directly infer likelihoods of all possible residue pairs in their entirety, which could effectively consider the correlations among multiple residues. We further construct the alignment with maximum likelihood, and finally build structure model according to the alignment. Tested on three independent datasets with in total 6,688 protein alignment targets and 80 CASP13 TBM targets, our method achieved much better alignments and 3D structure models than the existing methods including HHpred, CNFpred, CEthreader and DeepThreader. These results clearly demonstrate the effectiveness of exploiting the context-specific alignment motifs by deep learning for protein threading.

bioinformatics↗