Biology subjects

Hosokawa, S.

Publications and source records attributed to Hosokawa, S..

4 recordsLinked to original sources

bioRxiv2026

REN-former prioritizes candidate regulators of kidney disease-state transitions through single-cell foundation modeling and human genetics

Background Acute kidney injury (AKI)-to-chronic kidney disease (CKD) transition is associated with dynamic changes in tubular cell state. However, conventional single-cell transcriptomic analyses primarily identify genes differentially expressed between disease states and do not directly evaluate genes associated with directional transitions between cellular states. Methods We developed REN-former by fine-tuning Geneformer using the GSE183276 single-cell RNA-sequencing dataset from 45 participants representing Normal Reference, AKI, and CKD states. In silico gene deletion and overexpression analyses were used to estimate directional transcriptomic shifts, with a focus on proximal tubular cells. Selected genes were evaluated using summary-data-based Mendelian randomization (SMR), colocalization and expression analysis in additional KPMP participants not included in GSE183276. Results REN-former achieved recall values of 0.99, 0.80, and 0.79 for Normal Reference, AKI, and CKD, respectively. In silico perturbation analyses identified distinct gene programs associated with transitions from Normal Reference to AKI, from Normal Reference to CKD, from AKI to CKD, and from CKD to Normal Reference. Conventional analysis showed metabolic suppression and increased inflammatory and stress-response activation. SMR identified IFITM3, CALR, TTR, CALM1, MUC13, and RPL13, and colocalization supported IFITM3, CALR, TTR, and CALM1. In additional KPMP data, IFITM3 was higher, whereas TTR and CALM1 were lower, in CKD proximal tubules; CALR did not differ significantly. The observed expression changes were concordant with the REN-former-predicted directions for TTR and CALM1 but discordant for IFITM3. Conclusion REN-former provides a framework for prioritizing candidate regulators of kidney disease-associated cell states by integrating predicted perturbation effects with human genetic and transcriptomic evidence.

bioinformatics↗

bioRxiv2026

From introduction to eradication: reconstructing population size and removal history of an invasive species

O_LIUnderstanding the processes underlying successful eradication of invasive species is essential for achieving global island conservation goals. Despite the widespread availability of capture records from eradication programs, modeling frameworks that utilize these datasets to elucidate spatio-temporal population dynamics remain underdeveloped. C_LIO_LIIn this study, we reconstructed the spatio-temporal population dynamics of the small Indian mongoose on Amami-Oshima Island (712 km{superscript 2}), Japan, where the species was introduced in 1979 and officially declared eradicated in 2024 after more than 30 years of systematic removal. We integrated introduction records, capture data, and monitoring data using a hierarchical harvest-based model (HBM). To evaluate the models capacity to support management decisions and assess eradication success, we conducted retrospective analyses and compared estimated eradication probabilities with those obtained from a rapid eradication assessment (REA; Samaniego-Herrera et al., 2013). C_LIO_LIThe estimated population size (before reproduction) peaked at 5,449 individuals (95% CI: 4,703, 6,175) in 2000 and subsequently declined almost monotonically. The maximum invaded area was 547.78 km{superscript 2} (posterior median, 95% CI: 496.47, 566.04) in 2009, indicating that the removal program successfully prevented island-wide expansion. Retrospective analyses showed that population estimates remained within the 95% credible intervals of the full dataset estimates, demonstrating temporal consistency. Eradication probabilities estimated by the HBM were substantially higher than those from the REA, highlighting the sensitivity of estimates to fine-scale heterogeneity in detection processes. C_LIO_LISynthesis and applications: Hierarchical HBMs provide a powerful framework for reconstructing, predicting, and evaluating invasive species eradication dynamics. Being aware of the limitations for application to eradication evaluations, HBMs can support adaptive management in long-term eradication programs and improve our understanding of the mechanisms underlying successful eradication. C_LI

ecology↗

bioRxiv2026

The REFLEX system enables in vivo identification of perivascular angiogenic macrophages in the heart

Direct identification of physically interacting cells in vivo remains challenging because conventional interactome analyses infer signaling partners from transcriptomes and cannot reveal which cells are in direct contact. In pressure-overload induced cardiac remodeling, VEGF-A plays a central role in the maintenance of vascular integrity and cardiac function. However, the cell type which produces VEGF-A and how the VEGF-A peptide is delivered to vascular endothelial cells remains unclear. Here, we developed a genetically encoded platform that combines REFLEX mice with HUNTERuni-seq, enabling unbiased detection and transcriptional profiling of the cells that physically interact with vascular endothelial cells. The REFLEX and HUNTERuni-seq approach identified subpopulations of Vegfa positive macrophages which we named perivascular angiogenic macrophages (PVAMs). Although the amount of VEGF-A in PVAMs is small, loss of VEGF-A in PVAMs impaired angiogenesis and systolic function during pressure overload. We additionally show that direct contact between PVAMs and endothelial cells is critical for the delivery of VEGF-A to endothelial cells. Conventional interactome analysis predicted that cardiomyocytes as dominant sources of VEGF-A in the heart. However, cardiomyocyte Vegfa deletion had no effect on capillary density nor systolic function in a model of heart failure. These results suggest that VEGF-A signaling does not rely on free diffusion through the interstitium and that cellular proximity and physical contact between PVAMs and endothelial cells are the key determinants of effective signal delivery. Together, these findings establish REFLEX and HUNTERuni-seq as a versatile platform for uncovering biologically critical cell-to-cell interactions and provide new insight into intercellular communication in pathological tissue contexts.

cell biology↗

bioRxiv2025

PAH-former: Transfer Learning for Efficient Discovery of Pulmonary Arterial Hypertension-Associated Genes

Single-cell RNA sequencing (scRNA-seq) of patient samples holds promise for understanding disease mechanisms, but faces the challenge of excessive cost and effort in acquisition, processing, and data analysis, making it essential to leverage existing data. Pulmonary artery hypertension (PAH) is a refractory disease characterized by pulmonary vascular remodeling, and access to patient specimens is limited due to difficulties in tissue collection. In this study, we employed transfer learning with Geneformer, a deep learning algorithm pre-trained with scRNA-seq datasets and fine-tuned it with public PAH lung tissue data to identify the disease-relevant genes. The resulting algorithm, which we named PAH- former, demonstrated that its prediction accuracy varied significantly depending on the dataset used for fine-tuning. PAH-former enabled us to perform in silico perturbation analysis and identified PAH related genes. Loss-of-function PAH related genes in human pulmonary artery endothelial cells increased the expression of SOX18, a signature gene of PAH. This integration of artificial intelligence and biological experiments can significantly advance our understanding of molecular mechanisms of PAH.

bioinformatics↗