Bioengineering Center

Systematic clinical phenotyping using Human Phenotype Ontology (HPO) is central to rare disease diagnosis. However, current disease prioritization (ranking candidate diseases from HPO for a patient) methods face key challenges: they often fail to account for the hierarchical structure of HPO terms, ignore dependencies among correlated terms, and do not adjust for batch effects arising from systematic differences in phenotype documentation across cohorts, institutions, or clinicians. We aim to develop a scalable and statistically principled framework to address these limitations for rare disease prediction and patient stratification. We developed PhenoSS, a Gaussian copula-based framework that models disease-specific marginal prevalence of HPO terms while capturing their joint dependencies through a multivariate normal distribution. Phenotype frequencies were estimated using external curated resources, including OARD (Open Annotations for Rare Diseases) and HPO annotations. PhenoSS supports both pair-wise phenotype similarity calculation for patient clustering and posterior odds estimation for patient-specific disease prioritization. A batch-effect correction module mitigates systematic phenotyping differences across datasets. Across diverse simulation scenarios, PhenoSS demonstrated robust disease-prediction performance and consistently improved accuracy after batch-effect correction. In real electronic health record (EHR) data, PhenoSS identified clinically meaningful patient clusters and effectively distinguished patients with different rare diseases. In disease prioritization tasks, PhenoSS achieved competitive performance with existing methods, particularly for patients exhibiting sparse or noisy phenotype annotations. PhenoSS provides a statistically interpretable framework for modeling phenotypic heterogeneity in rare disease research and is adaptable to other structured clinical vocabularies such as SNOMED-CT and ICD codes.

This study investigates the effects of three different drying methods - sun drying, shade drying and oven drying at 55 °C. Some physical, structural and chemical properties of Ziziphus jujuba Mill. fruit before and after drying are evaluated. Freshly harvested jujube fruits from the Elazığ (Turkey) region are subjected to each drying method, and changes in surface hardness, moisture content and internal structure are analyzed. Shore A hardness measurements indicate a significant increase in fruit firmness over time, with oven-dried samples showing the highest values, followed by sun-dried samples. Fourier transform infrared spectroscopy reveals marked chemical transformations during drying, including reductions in OH stretching bands and shifts in CO and CH regions, reflecting moisture loss and modifications in sugars and polysaccharides. Weight loss calculations determine that approximately 60-65 wt% of initial mass is lost to reach a final moisture content of around 16-18 wt%. These structural and chemical changes influence the fruit's texture, rehydration ability and consumer acceptability. Drying kinetic modeling using two-term models for both oven and sun/shade drying demonstrates excellent fit (R2 = 0.9995), with oven drying showing a higher effective diffusion coefficient (Deff = 4.57 × 10-10 m2 s-1) than sun/shade drying (Deff = 7.96 × 10-11 m2 s-1), indicating more efficient moisture transport. Response surface methodology is employed to optimize drying conditions and evaluate mass loss over time. © 2026 Society of Chemical Industry.

While induced pluripotent stem cells (iPSCs) have gained popularity in studying neurodegenerative diseases, the heterogeneity of stem cells used across studies impacts cross-study comparison. The iPSC Neurodegenerative Disease Initiative (iNDI) selected the KOLF2.1J cell line and prioritized its use as a reference standard for studying the effects of pathogenic variants on cell biology due to its stability and neutral neurodegenerative disease genetic risk. This cell line, and its derivatives expressing over 100 variants related to Alzheimer’s disease, Parkinson’s disease, and other neurological diseases, are available for academic and industry access. Current genomic data analyses are limited by the use of a human reference genome that does not capture the complete genetic background of a given iPSC line. While in the future this issue may be partially mitigated by the creation of a comprehensive human pangenome, previous work has shown that generating custom genomes is of value both to characterize the variation present and to serve as a more appropriate genomic reference. Here, we generated and characterized a custom complete genome assembly from KOLF2.1J. Mapping of sequencing reads to a personalized diploid assembly results in more comprehensive mapping compared to traditional linear references (i.e GRCh38). In addition, we provide a comprehensive custom gene annotation along with isoform expression and differential methylation analyses across multiple cell types. The assembly and all additional data is browsable and publicly available. This resource will enable more accurate investigation of the KOLF2.1J cell line and any genomics data generated compared to using traditional generalized references, while also serving as a foundational approach for establishing custom reference assemblies for other high-value iPSC lines.

Radiation modulation is critical due to the growing risks of radiation exposure in medical, occupational, and environmental settings. This study explored the potential of heterogeneous nuclear ribonucleoproteins (hnRNPs), particularly hnRNP A2B1, as therapeutic targets for radiomodulatory drugs. In vitro experiments on human endothelial cells exposed to various X-ray doses revealed a dose-dependent increase in hnRNP A2B1 expression, with the histochemical score increasing significantly from 105 to 280 at 8.0 Gy. Using in silico molecular docking, 33 radiomodulatory agents were evaluated for their binding affinities with hnRNP A2B1. Notable ligands, such as hesperidin and psoralidin, demonstrated strong binding affinities (ΔGtot values: −17.2 and −17.9 kcal⋅mol-1, respectively). Finally, molecular dynamics simulations confirmed that psoralidin had the highest affinity for hnRNP A2B1 using implicit solvation models. Overall, this study revealed that hnRNP A2B1 is vital for cellular radiomodulation and is a promising target for radiomodulatory drugs that could have radioprotective or radiosensitizing effects.

This review focuses on current challenges associated with the study and standardization of chitosan, a promising biopolymer for medical applications. Its key properties, such as biodegradability, low toxicity, mucoadhesion, and antimicrobial effects, make it highly sought in various fields of medicine. Particular attention is given to the analysis of chitosan parameters, including molecular weight, degree of deacetylation, and pattern of acetylation, as well as compliance with pharmacopoeial requirements. The challenges associated with the reproducibility of chitosan properties and the absence of uniform analytical standards are addressed. The review also summarized the data on the relationship between the chitosan structural characteristics and biological activity (antimicrobial, antioxidant, immunomodulatory, etc.), which is important for predicting its behavior in biological systems. Regulatory considerations governing the medical use of chitosan and prospects for its introduction into medical practice are examined. This review will be useful for researchers engaged in the development, characterization, and standardization of chitosan-based biomaterials.

Chiral β-cyano ester scaffolds are highly valuable synthons in pharmaceutical synthesis. The direct asymmetric hydrogenation of β-cyanoacrylates, catalyzed by ene-reductases (ERs), offers an efficient route to these scaffolds with a high atom economy under mild reaction conditions. However, the narrow substrate acceptance of ER reported in the previous work has limited its application in asymmetric hydrogenation. In this study, we engineered the substrate-binding pocket of an ER from Saccharomyces eubayanus (SeER) to access a panel of chiral β-cyano ester scaffolds with broad structural diversity. Using residue-deletion engineering, mutant M1 was generated and exhibited 416-fold higher catalytic activity toward the model substrate compared to the wt-SeER, along with improved enantioselectivity (from rac to 97% ee). The optimal mutant M4, obtained through saturation mutagenesis and iterative combination, exhibited 1.36 × 104-fold higher catalytic efficiency than that of wt-SeER, and the enantioselectivity further increased to 99.7% ee. Mutant M4 demonstrates broad substrate acceptance and was successfully applied in a gram-scale (200 mM) asymmetric synthesis of Pregabalin (a first-line antiepileptic agent) with 99.7% ee and 65% isolated yield. This study highlights residue deletion as an effective strategy for modifying the ER binding pocket to access the asymmetric hydrogenation of β-cyanoacrylates but also provides valuable engineering guidance for enhancing the substrate acceptance of other ERs possessing similar pockets.

Accurate delineation of striatal subregions in brain MRI is essential for reliable dopaminergic PET quantification and the assessment of neurodegenerative disorders such as Parkinson’s disease. In this study, we introduce StriaSeg-iARM, a cascaded deep learning framework designed for high-precision segmentation of 12 anatomically defined striatal subregions in native space. The model employs a two-stage 3D residual U-Net architecture, where the first stage localizes the striatum and the second performs voxel-wise subregional classification. To improve segmentation accuracy in anatomically altered cases, the framework incorporates an intermediate refinement step, enabling expert correction of the striatal mask prior to final inference. The model was trained on multi-scanner T1-weighted MRI data and evaluated on both internal and external datasets, including real-world cases with striatal atrophy. Compared to atlas-based normalization, a conventional single-stage model, and a cascaded model without refinement, StriaSeg-iARM achieved superior segmentation performance, with an average Dice coefficient of 0.91. Furthermore, segmentation accuracy was enhanced through expert intervention without requiring model retraining. Quantitative PET analyses based on StriaSeg-iARM also exhibited improved agreement with expert-defined segmentations. Overall, these results demonstrate the practical applicability of StriaSeg-iARM as a flexible and extensible framework for subregional segmentation and PET quantification.

Potato tubers can be dormant for 1–15 weeks after harvest, and the germination of tuber eye-buds is suppressed by cold. Genetic mechanisms regulating bud dormancy at low temperatures are not completely understood. We performed RNA-seq to compare gene expression in tubers after 0.5, 3.5, and 6.5 months of cold storage. Differentially expressed genes (DEGs) in non-dormant (3.5/6.5 months) vs dormant (0.5 months) tubers were associated with transcriptional and translational activation, cell growth, metabolism, hormonal signaling, meristem development, dormancy break, and reproduction, confirming the non-dormant state of the meristem in the middle of storage. DEGs encoding transcription factors (TFs) (CBF1, FLC, SVP, HY5, GI, CO, FT, SOC1, CDF1, POTM1) were associated with the regulation of dormancy, flowering, and tuberization. TF DEGs upregulated (78) or downregulated (224) over four times in non-dormant vs dormant tubers are considered potential coordinators of the endo- to non-dormancy transition and upcoming tuber sprouting during cold storage. RT-qPCR analysis of non-dormant and germinating buds revealed the upregulation of tuberigens (StSP3D, StFTL1-1, StFTL1-2), anti-tuberigens (StSP5G-like), and TF genes positively associated with tuberization or flowering/germination (StFDL1, StFDL, StCDF1, StCO-like). Our results should further investigation of the mechanisms underlying tuber meristem dormancy release and sprouting during long-term cold storage.