Biology subjects

Oh, T.-J.

Publications and source records attributed to Oh, T.-J..

7 recordsLinked to original sources

bioRxiv2025

Prediction of bacterial protein-compound interactions with only positive samples

Prediction of Compound-Protein Interactions (CPI) in bacteria is crucial to advance various pharmaceutical and chemical engineering fields, including bio-catalysis, drug discovery, and industrial processing. However, current CPI models cannot be applied for bacterial CPI prediction due to the lack of curated negative interaction samples. This paper introduces a novel Positive-Unlabeled (PU) learning framework, named BIN-PU, to address this limitation. BIN-PU generates pseudo positive and negative labels from known positive interaction data, enabling effective training of deep learning models for CPI prediction. We also propose a weighted positive loss function that weights to truly positive samples. We have validated BIN-PU with multiple CPI backbone models, comparing the performance with the existing PU model using bacterial cytochrome P450 (CYP) data. Extensive experiments demonstrate the superiority of BIN-PU over the benchmark model in predicting CPIs with only truly positive samples. Furthermore, we have validated BIN-PU on additional bacterial proteins obtained from literature review, human CYP datasets, and uncurated data for its reproducibility. We have also validated the CPI prediction for the uncurated CYP data with biological and biophysical experiments. BIN-PU represents a significant advancement in CPI prediction for bacterial proteins, opening new possibilities for improving predictive models in related biological interaction tasks.

bioinformatics↗

bioRxiv2025

Plastic-hydrolytic enzyme classification using explainable deep learning

The rapid accumulation of plastic waste has emerged as a critical environmental threat, driving the need for scalable and effective biodegradation solutions. Hydrolytic plastic-degrading enzymes (PDEs) offer a promising solution, yet their functional classification remains limited by insufficient annotations and enzymatic diversity. In this study, we present an explainable deep learning framework, PEPIC, to classify nine types of PDEs directly from protein sequences. Using a curated dataset of experimentally validated enzymes and an expanded homologous dataset, we built an explainable deep learning model based on convolutional neural networks (PEPIC) for plastic-degrading enzyme prediction. We benchmarked PEPICs performance against state-of-the-art approaches. First, PEPIC demonstrated statistically significant improvements in predictive performance compared to state-of-the-art methods. Second, PEPIC calculates contribution scores for each amino acid in the protein sequence, indicating their influence on the predictions. The model interpretation revealed that regions highlighted by high contribution scores matched conserved catalytic triads and substrate-binding clefts across PET-, PCL-, and PLA-degrading enzymes. Furthermore, structural modeling confirmed the trustworthiness of PEPICs predictions. Finally, PEPIC predicted an uncurated enzyme as a PET-degrading enzyme, which was biologically validated to hydrolyze bis(2-hydroxyethyl) terephthalate (BHET). These findings demonstrate that PEPIC provides accurate and trustworthy predictions of PDEs, facilitating the discovery of novel enzymes and supporting the development of sustainable plastic biodegradation technologies.

bioinformatics↗

bioRxiv2025

Resolving Oligomeric States of Photoactivatable Proteins in Live Cells via Photon Counting Histogram Analysis

Oligomerization of photoactivatable proteins is widely used in optogenetics to modulate protein activity and regulate biological processes. However, their oligomerization states remain challenging to quantify in living cells. We applied photon counting histogram (PCH) analysis to quantify the oligomerization of two commonly used photoactivatable proteins, Vaucheria frigida Aureochrome light-oxygen-voltage (VfAuLOV) and Arabidopsis Thaliana cryptochrome 2 (AtCRY2), under dark and light conditions, providing a direct measurement of oligomerization states in live cells. Under blue light stimulation, VfAuLOV primarily forms dimers, whereas AtCRY2 forms higher order multimers in HEK293T cells. At intracellular concentrations above 1000 nM, AtCRY2 transitions into tetramers upon light stimulation, consistent with structural data obtained from cryo-electron microscopy. Unexpectedly, AtCRY2 (W374), a constitutively active mutant, still exhibits light-induced oligomerization. Human cryptochrome 2, in contrast, shows light-independent oligomerization at the intracellular concentration above 100 nM, below which the monomeric state dominates. Optogenetic signaling outcomes, demonstrated by light-induced lytic cell death, align with the oligomerization state of tested proteins. This study presents a quantitative framework for elucidating protein oligomerization in live cells, thereby enhancing the understanding of optogenetic mechanisms and their role in cellular signaling. Significance StatementPhotoactivatable proteins enable the development of optical actuators that modulate cell signaling through light-mediated protein-protein interaction. Light-oxygen-voltage (LOV) and cryptochrome 2 (CRY2) are commonly used photoactivable protein modules that undergo blue light-mediated oligomerization, and the latter has been used to create synthetic biomolecular condensates in live cells. Although in vitro structural studies suggested their oligomerization states, their behavior in live cells is unclear. Here, we applied photon counting histogram analysis to determine the dark and light-activated oligomerization states of these photoactivatable proteins in live cells. Insights gained will provide better guidance in designing optical actuators, leading to more quantitative interpretation of the light-mediated signaling outcomes in cells.

biophysics↗

bioRxiv2025

HIT-EC: Trustworthy prediction of enzyme commission numbers using a hierarchical interpretable transformer

Accurate and trustworthy prediction of Enzyme Commission (EC) numbers is critical for understanding enzyme functions and their roles in biological processes. Despite the success of recently proposed deep learning-based models, there remain limitations, such as low performance in underrepresented EC numbers, lack of learning strategy with incomplete annotations, and limited interpretability. To address these challenges, we propose a novel hierarchical interpretable transformer model, HIT-EC, for trustworthy EC number prediction. HIT-EC employs a four-level transformer architecture that aligns with the hierarchical structure of EC numbers, and leverages both local and global dependencies within protein sequences for this multi-label classification task. We also propose a novel learning strategy to handle incomplete EC numbers. HIT-EC, as an evidential deep learning model, produces trustworthy predictions by providing domain-specific evidence through a biologically meaningful interpretation scheme. The predictive performance of HIT-EC was assessed by multiple experiments: a cross-validation with a large dataset, a validation with external data, and a species-based performance evaluation. HIT-EC showed statistically significant improvement in predictive performance when compared to the current state-of-the-art benchmark models. HIT-ECs robust interpretability was further validated by identifying well-known conserved motifs and functional regions in the CYP106A2 enzyme family. HIT-EC would be a robust, interpretable, and reliable solution for EC number prediction, with significant implications for enzymology, drug discovery, and metabolic engineering. The open-source code is publicly available at: https://github.com/datax-lab/HIT-EC.

bioinformatics↗

bioRxiv2025

Interpretable Kolmogorov-Arnold Networks for Enzyme Commission Number Prediction

Accurate prediction of enzyme commission (EC) numbers remains a significant challenge in bioinformatics, limiting our ability to fully understand enzyme functions and their roles in biological processes. This paper presents the integration and evaluation of the next paradigm of deep learning architecture, named Kolmogorov-Arnold network (KAN), in state-of-the-art models for predicting EC numbers. KAN modules are incorporated into current state-of-the-art models to assess their impact on predictive performance. Additionally, we introduce a novel interpretation method, specifically designed for KANs, to identify relevant input features for a given prediction, addressing a current limitation in KANs. Our evaluation demonstrates that the integration of KANs significantly enhances predictive performance compared to the state-of-the-art deep learning models, with up to a 4.35% increase in micro-averaged F1 score and a 4.1% increase in macro-averaged F1 score. Moreover, our novel interpretation method not only enhances the predictions trustworthiness but also facilitates the discovery of motif sites within enzyme sequences. This innovative approach provides deeper insights into enzyme functionality and highlights potential new targets for research. The results underscore KANs effectiveness in improving enzymatic classification and advancing our understanding of enzyme structures and functions. The open-source code is publicly available at: https://github.com/datax-lab/kan_ecnumber.

bioinformatics↗

bioRxiv2020

Genome-Enabled Discovery of Anthraquinone Biosynthesis in Senna tora

Senna tora is a widely used medicinal plant. Its health benefits have been attributed to the large quantity of anthraquinones, but how they are made in plants remains a mystery. To identify the genes responsible for plant anthraquinone biosynthesis, we sequenced and annotated the genome of S. tora at the chromosome level with contig N50 and super-scaffold N50 of 4.03 Mb and 41.7 Mb. Comparison among related plant species showed that a chalcone synthase-like (CHS-L) gene family has lineage-specifically and rapidly expanded in S. tora. Combining genomics, transcriptomics, metabolomics, and biochemistry, we identified a CHS-L responsible for biosynthesis of anthraquinones, the first example in plants. The S. tora reference genome will accelerate the discovery of biologically active anthraquinone biosynthesis pathways in medicinal plants. One Sentence SummaryThe chromosome-scale reference genome of a medicinal plant Senna tora, transcriptomics, metabolomics, and biochemical analysis provide new insights into anthraquinone biosynthesis in plants.

plant biology↗

bioRxiv2019

De novo transcriptome sequence of Senna tora provides insights into anthraquinone biosynthesis

Senna tora is an annual herb with rich source of anthraquinones that have tremendous pharmacological properties. However, there is little mention of genetic information for this species, especially regarding the biosynthetic pathways of anthraquinones. To understand the key genes and regulatory mechanism of anthraquinone biosynthesis pathways, we performed spatial and temporal transcriptome sequencing of S. tora using short RNA sequencing (RNA-Seq) and long-read isoform sequencing (Iso-Seq) technologies, and generated two unigene sets composed of 118,635 and 39,364, respectively. A comprehensive functional annotation and classification with multiple public databases identified array of genes involved in major secondary metabolite biosynthesis pathways and important transcription factor (TF) families (MYB, MYB-related, AP2/ERF, C2C2-YABBY, and bHLH). Differential expression analysis indicated that the expression level of genes involved in anthraquinone biosynthetic pathway regulates differently depending on the degree of tissues and seeds development. Furthermore, we identified that the amount of anthraquinone compounds were greater in late seeds than early ones. In conclusion, these results provide a rich resource for understanding the anthraquinone metabolism in S. tora.

plant biology↗