Biology subjects

Datta, A.

Publications and source records attributed to Datta, A..

At least 19 recordsLinked to original sources

bioRxiv2026

Human-specific remodeling of an endogenous retrovirus shapes structural diversity at acrocentric nucleolar organizer regions

Human acrocentric short arms harbor K111, an endogenous retrovirus that became an integral component of nucleolar organizer region (NOR)-associated architecture across all five acrocentric chromosome types. Here we combine complete human and primate genomes, haplotype-resolved population assemblies and parent-offspring trios to reconstruct its evolution and transmission. Predominantly full-length ancestral elements underwent human-specific expansion and remodeling, generating recurrent mosaics and individual-specific multicopy configurations. Population structural variation is concentrated in surrounding satellite landscapes rather than within K111-derived sequences themselves. Trio-resolved genomes reveal direct parental transmission, single-parent remodeling and, crucially, maternal-paternal remodeling in distinct configurations: large reciprocal domains with contrasting parental affinities and de novo sequence exchange within an offspring locus. Raw long reads support both maternal-paternal configurations. K111-derived loci associate with nucleolin-positive compartments. Together, these findings establish K111 as an evolutionarily dynamic component and molecular record of acrocentric NOR architecture, revealing unrecognized remodeling and sequence generation across human generations.

genetics↗

bioRxiv2026

Model-based assessment of race, sex, and electrode montage in ECT

Objective: This computational modeling study quantified the influence of sex and race-related cranial anatomy on predicted brain-wide current flow during electroconvulsive therapy (ECT) across conventional (bifrontal (BF), bitemporal/bilateral (BL), right unilateral (RUL)) and experimental (focal electrically administered seizure therapy (FEAST) and frontomedial (FM)) electrode montages. The objective was to determine whether race-associated variability meaningfully contributes to differences in ECT stimulation metrics across montages. Methods: Finite element head models of Chinese, Black, and Caucasian subjects were developed using high-resolution magnetic resonance imaging and analyzed using the Realistic vOlumetric- Approach-based Stimulator for Transcranial electric stimulation (ROAST) pipeline (N = 150 total; n = 50 per cohort, comprising 25 M and 25 F, age range: 20-30 years). Five ECT montages were simulated under a constant-current condition (900mA). Stimulation strength (Ebrain/Eth) was quantified as 90th percentile of brain-wide E-field magnitude (Ebrain) relative to neuronal activation threshold (Eth = 0.25 V/cm) quantified stimulation strength. Overall focality was evaluated as a percentage of brain volume stimulated above the neural activation threshold (Ebrain [≥] Eth), while laterality was quantified as the median right-to-left hemispheric E-field magnitude ratio. The effects of race, sex, and montage on stimulation strength, focality, and hemispheric laterality were statistically analyzed. Results: Substantial race- and sex-related differences observed in cranial anatomy resulted in systematic variation in predicted ECT-induced E-field intensity. Brain-wide E-field magnitude varied by both race and montage, with the largest fields generally observed in Caucasian head models and during BL stimulation. Montage exerted the strongest effect on stimulation strength (Ebrain/Eth) with BL and FEAST producing the highest stimulation strengths, followed by RUL and FM, while BF produced the lowest. Caucasian subjects generally predicted higher stimulation strengths than Black and Chinese subjects, whereas females predicted modestly higher stimulation strengths than males. Laterality was primarily determined by montage, with FEAST producing the greatest hemispheric asymmetry, followed by RUL. Chinese subjects demonstrated higher laterality ratios than both Black and Caucasian subjects. BL, RUL, and FEAST stimulated substantially larger brain volumes above neural activation threshold (less focal stimulation) than BF. Lower focality was observed in Caucasian subjects relative to Black and Chinese subjects, and in females relative to males. Conclusions: Electrode montage was the primary determinant of predicted ECT stimulation strength, focality, and laterality. Race-related anatomical differences and, to a lesser extent, sex-related differences systematically altered stimulation patterns, supporting consideration of individualized anatomy in ECT dosing and treatment optimization.

neuroscience↗

bioRxiv2026

Mechanism of K63-linked polyubiquitin recognition and cleavage by the BRCA1-A complex

Deubiquitylases modulate cellular processes by cleaving monoubiquitin or polyubiquitin chains. The ARISC-RAP80 complex partners with BRCA1-BARD1 to form the BRCA1-A super-complex, which recognizes K63-linked ubiquitin chains at DNA damage sites. ARISC-RAP80 contains multiple ubiquitin-binding sites, yet how these influence recognition and cleavage of K63-polyubiquitylated substrates remains unknown. We discover that a composite three-subunit interface allows ARISC-RAP80 to position K63-linked polyubiquitin chains in its catalytic site. Substrate recognition is further supported by RAP80 and non-catalytic ubiquitin-binding sites that impose a compact conformation to K63-polyubiquitylated substrates. This mechanism exploits the inherent flexibility of long ubiquitin chains and differs considerably from other deubiquitylases. Structure-guided mutagenesis validate ubiquitin chain interactions, and cell-based assays demonstrate a functional role of the observed interfaces in chromatin recruitment. Our findings define mechanisms of polyubiquitin chain decoding and cleavage by ARISC-RAP80, linking ubiquitin reading and erasing functions to BRCA1-A mediated DNA damage responses.

biochemistry↗

bioRxiv2026

Governance scale and network structure shape pollinator recovery under pesticide reduction

Reducing pesticide risks while maintaining food production remains a central challenge for sustainable agriculture. Although pesticide reduction is pursued through centralized regulation and farm-level Integrated Pest Management, how these governance pathways translate into pollinator recovery in agroecological systems remains poorly understood. Existing ecological network models often treat pesticide pressure asm external forcing and management actions as fixed parameters, limiting their ability to capture feedbacks among governance decisions, network structure, and population dynamics. Here, we develop a dynamical framework that embeds pesticide management within tripartite pollinator-plant-pest networks using a policy variable and a farm-level adoption variable. Across empirical and synthetic networks, we show that recovery is not determined by pesticide reduction alone, but by how management acts through ecological interaction structure. More modular networks require stronger intervention, and pollinators with similar degrees show different recovery outcomes, indicating that degree alone does not determine recovery potential. Further, increasing policy strength generally expands the persistence domain more than increasing farmer adoption alone. These results show that pesticide reduction does not automatically yield ecological recovery, and effective strategies must match governance scale to ecological condition and network structure.

ecology↗

bioRxiv2026

Phenological Plasticity without Directional Change: Tropical Hornbills Track Temperature but Maintain Reproductive Stability

Climate change is altering breeding phenology in birds worldwide, yet tropical systems remain understudied despite harbouring the greatest avian diversity. We drew upon 26 years (1998-2024) of breeding phenology and reproductive success data on three sympatric Asian hornbill species (Great Hornbill Buceros bicornis, Wreathed Hornbill Rhyticeros undulatus, Oriental Pied Hornbill Anthracoceros albirostris) from a tropical forest in northeast India to examine whether trends in long-term breeding parameters reflect directional climate change or interannual variability. Over the study period, we found no significant directional trends in nest entry timing, occupancy, or breeding success over 26 years, despite modest regional warming. However, breeding parameters showed sensitivity to interannual temperature variation, with pre-breeding temperature (January-February) explaining 36-70% of variance in nest timing. Temperature sensitivity differed among species, with Wreathed Hornbill showing the strongest response (-15.9 days/{degrees}C), followed by Great Hornbill (-9.8 days/{degrees}C), and Oriental Pied Hornbill showing the weakest (-4.8 days/{degrees}C), consistent with predictions that larger, frugivorous species with extended breeding cycles would exhibit greater climatic sensitivity. The underlying mechanism appears related to a body size-phenology-thermal exposure interaction: Oriental Pied Hornbills nest later (mid-April) with shorter breeding cycles ([~]90 days), completing reproduction before peak monsoon heat (June), whereas both larger species nest earlier (mid-February) but endure thermal stress throughout extended breeding cycles ([~]130+ days) into August. Notably, annual occupancy and breeding success remained stable across years despite strong phenological responses to temperature, suggesting that phenological plasticity may help maintain reproductive performance under current climatic variation. These findings indicate that tropical cavity-nesting frugivores exhibit phenological plasticity coupled with demographic stability, though the mechanisms underlying this apparent buffering remain incompletely understood, and body size-dependent differences in thermal exposure suggest heterogeneous vulnerabilities to future warming scenarios.

ecology↗

bioRxiv2026

Metal homeostasis is remodeled in response to different quiescence triggers

Cells can enter a reversible non-proliferative state called quiescence either spontaneously or in response to nutrient deprivations. Metal ions are essential nutrients and play wide-ranging regulatory and signaling roles in biological processes. We previously showed that zinc is an essential nutrient for the mammalian cell cycle as zinc deprivation drives cells into quiescence, and this quiescent state is associated with changes in iron, copper, and manganese, suggesting broad remodeling of metal homeostasis. Here we examine whether metal remodeling is a general feature of quiescence by inducing quiescence via different triggers (zinc deficiency, serum starvation, and growth factor withdrawal) in MCF10A cells. Fluorescence microscopy and elemental analysis reveal significant trigger-dependent changes in the labile and total metal pools of quiescent cells. To gain insight into these differences, we carried out RNA sequencing and differential expression analysis, focusing on metal associated, metal regulatory, and metal homeostasis genes. While core quiescence pathways are shared across triggers, quiescence states remain molecularly distinct. A significant percent of 2458 metal homeostasis annotated genes are differentially expressed including 55% in starvation-induced quiescence, 50% in zinc deficiency-induced quiescence and 21% in growth factor withdrawal-induced quiescence. Our results also showed unique alteration of genes involved in major metal dependent processes including antioxidant activity, oxidative phosphorylation, heme metabolism, and chromatin accessibility in different quiescence states. Overall, this work demonstrates that metal homeostasis is systematically rewired during cellular quiescence with associated effects on genes that regulate critical biological processes. Significance StatementCells constantly evaluate their nutrient and energy status and integrate these signals into proliferation-quiescence decisions. Quiescence prevents cells from passing damage to daughter cells. Metals are essential micronutrients for biological processes. While limitation of zinc can drive cells into quiescence and alter other metals, how metals are remodeled and whether this is a common feature of quiescence was unknown. Here we examined three quiescence triggers by modifying growth media and serum, the primary source of metals. We report significant changes in labile and total Cu, Fe, Zn and Mn pools, and metal associated genes in response to distinct quiescence triggers. These changes converge on mitochondrial function, cellular antioxidant activity, and heme biosynthesis in a trigger specific manner.

cell biology↗

bioRxiv2026

Landscape of reversion alterations in homologous recombination genes reveals evolutionary constraints

Reversion mutations (REVs) restore homologous recombination repair (HRR) and confer resistance to PARP inhibitors (PARPi) in HR-deficient cancers. Yet, their prevalence, mechanisms, and biological constraints remain undefined. We analyzed genomic profiling of 609,464 tissue and liquid biopsy samples across multiple cancer types to delineate the pan-cancer landscape of REVs. REVs were identified in eight HRR genes, most frequently BRCA2 and BRCA1 and notably never in ATM or CHEK2. REVs exclusively impacted truncating pathogenic variants, predominantly through large in-frame and exon-level deletions, associated with repetitive sequences. Conserved functional domains are relatively depleted of REVs. Structural modeling and functional studies support that exon-level deletions preserve critical domain architecture and confer PARPi resistance. The study establishes HRR reversion as a structurally permissive, yet evolutionarily constrained, resistance mechanism with implications on response, monitoring, and therapeutic strategy. One Sentence SummaryHomologous recombination reversion evolves through structurally tolerated, microhomology-driven alterations that restore DNA repair under therapeutic selection across cancers.

cancer biology↗

bioRxiv2026

The BRCA1-A complex restricts replication fork reversal-dependent DNA repair in ATM deficient cells

Ataxia Telangiectasia Mutated (ATM) kinase deficiency results in cancer susceptibility and drug sensitivity. Deficiency in either the BRCA1 interacting A complex or XRCC4/Ligase 4 confers resistance to Topoisomerase I or PARP1 inhibitors in ATM-deficient cells. This suggests that BRCA1-A directs toxicity to fork-damaging agents in ATM mutated cells vis-a-vis illegitimate end-joining. Here, we show that ATM inhibition triggers combined SUMO and ubiquitin mediated BRCA1-A damaged fork recognition to restrict end-resection and cause Topoisomerase I inhibitor hypersensitivity. BRCA1-A deficient cells display elevated chromatin accessibility and nuclease activity at damaged forks, coupled with restored resection and drug resistance. Electron microscopy evidence demonstrates that ATM inhibition prevents replication fork reversal, which is restored by BRCA1-A loss to generate substrates for end resection. These findings reveal that BRCA1-A enforces a restrictive chromatin state to suppress the genesis of resection substrates, implicating fork reversal as a key determinant of chemotherapy response in ATM deficient cells.

cell biology↗

bioRxiv2026

Climate-linked divergence in tree flowering and fruiting in an Eastern Himalayan tropical forest

PremiseLong-term phenological patterns are increasingly being examined from the perspective of climate change and its potential effects. Climatic effects on plant phenology could involve the direct responses to changes in temperature, precipitation and solar irradiance, or could be mediated by these variables through exogenous teleconnections such as the El Nino Southern Oscillation (ENSO). The effects of climatic fluctuations on inter-annual variation in tropical phenology remain understudied. MethodsWe examined long-term patterns of tree flowering and fruiting intensity in a tropical forest site in the Eastern Himalayas between 2011 and 2024. Species-specific patterns were examined for 36 species. Long-term patterns were quantified using Generalized Additive Models, and splines were visualized to infer trends. Through Generalized Linear Mixed Models, we determined if there was a lagged phenological response to ENSO and temperature, precipitation and solar irradiance, and whether ENSO effects were being mediated through the latter group of variables or plant traits. ResultsBetween 2011 and 2019, trends in flowering and fruiting were significant for 17 and 23 species respectively. Flowering increased for 7 species, while fruiting declined for 8 species. Flowering peaked during El Nino, but this association did not appear to be mediated through climate variables, whereas fruiting showed a three-month positive lagged response to solar irradiance, independent of ENSO. The peak season of reproduction was the only trait determining species-specific responses to climate variables. ConclusionOur study highlights nonlinearity in long-term patterns of reproductive phenology, and the importance of solar irradiance in determining inter-annual fruit production.

ecology↗

bioRxiv2026

Factors shaping frugivory patterns of Asian mammals using a continental-scale dataset

O_LIFrugivores vary in their selection of fruit traits and their fruit handling methods, leading to differences in the plant species they consume for fruits. While fruit consumption patterns of birds are relatively well understood, much less is known about those of mammals. C_LIO_LIGiven the wide morphological and physiological diversity of mammals, fruit consumption patterns and fruit traits selected by different mammal groups may vary substantially. C_LIO_LIWe investigated differences in fruit consumption among three mammal groups - primates, herbivores, and carnivores - in Asia based on peer-reviewed and secondary literature. We assessed both morphological traits and taxonomic composition and compared patterns across vegetation types and for figs and non-figs. We found that primates (29%) and carnivores (21%) consumed more unique fruit genera than herbivores (6%). Carnivores and primates shared more fruit genera with each other (17%) than with herbivores. These patterns were consistent across vegetation types and for figs and non-figs. C_LIO_LIMorphological traits such as fruit size, colour, type, habit, seed number, and seed arillation showed no major differences among mammal groups. C_LIO_LIThere was no significant relationship between mammal body size and the mean or maximum fruit diameter consumed. However, among mammals that handle fruits exclusively with their mouthparts, body size was positively related to the maximum fruit diameter consumed. In contrast, for mammals that handle fruits using opposable thumbs (primates), body size showed a negative relationship with the mean fruit size consumed. There was no significant relationship between mammal activity patterns and the colour of the fruits they consumed. C_LIO_LIOur results suggest that fruit consumption patterns among mammal groups are not strongly differentiated by the morphological traits investigated; however, carnivores and primates are more similar in their preferred fruit genera. Moreover, morphological trait selection may be influenced by fruit handling methods. C_LIO_LIHerbivores consumed larger fruits and, like primates, preferred dull-coloured fruits, whereas carnivores more often fed on liana and shrub fruits across a wider colour range C_LIO_LIFuture research should focus on chemical and quantitative visual traits, such as volatile profiles and nutritional composition, to better understand the drivers of mammal fruit consumption. C_LI

ecology↗

bioRxiv2026

LinkDTI: Drug-Target Interactionsprediction through a Link Predictionframework on Biomedical KnowledgeGraph

Computational drug-target interactions (DTI) prediction serves as a valuable tool for drug discovery and repurposing by cost-effectively narrowing down the potential drug-target space. This paper presents LinkDTI, a computational framework that predicts DTIs by identifying connections within a heterogeneous knowledge graph of drugs, proteins, diseases, and side effects. Unlike methods that rely on mathematical techniques like matrix completion or similarity-based scoring, LinkDTI uses an advanced graph-based approach to capture relationships between biomedical entities. Specifically, LinkDTI applies a modified version of the multilayer GraphSAGE model that learns from the heterogeneous knowledge graph and predicts potential drug-target interactions. Our model incorporates negative sampling that balances the data to address the issue of having more negative than positive interactions. Our results show that LinkDTI consistently performs better in AUROC and AUPRC than baseline methods by at least 2.5% across different sampling ratios and conditions. Subsequently, it identifies approximately 945 new potential DTIs, marking a 49.14% increase over known DTIs. Overall, LinkDTI offers a simple yet effective method for integrating diverse biomedical data to identify potential drug-target interactions. The code and data can be found at https://github.com/hub2nature/LinkDTI_heterogenous_KG.git.

bioinformatics↗

bioRxiv2026

Neuronavigation-free and MRI-free Localization of Deep Brain Targets

Emerging non-invasive brain stimulation modalities, including transcranial focused ultrasound and transcranial interferential stimulation, offer the promise of safely and non-invasively modulating deep brain structures. Accurate targeting of these regions typically involves subject-specific MRI, limiting widespread deployment. Here, we introduce an MRI-free, neuronavigation-free method for localizing deep brain targets using only three simple scalp measurements. These measures define an affine transformation that maps the MNI152 standard head model to an individuals head geometry. We evaluated our approach on 50 healthy adults, comparing our model{square}predicted coordinates against ground-truth coordinates obtained via MRI-based nonlinear normalization. Across ten deep brain targets, our method achieved a mean localization error of 3.82 mm demonstrating a more accessible and cost-efficient alternative than MRI- or neuronavigation-based approaches.

neuroscience↗

bioRxiv2026

Morphological bias of the MNI152 brain

The MNI152 template is widely treated as a representative average brain in neuroimaging, computational modeling, and neuromodulation research, yet its fidelity to true population morphology has not been systematically evaluated. In this study, we compared the MNI152 template to 430 individual MRI scans from a publicly available dataset spanning Asian, Black, and White participants. We additionally generated an alternative template using ANTsPy to assess whether a modern diffeomorphic approach yields a more anatomically representative average. We conducted affine registrations and deformation-based morphometry to detect and quantify gross morphological differences as well as local voxel-level deformations. Across all racial groups, the MNI152 template required consistent global contraction, and its Jacobian fields revealed spatially heterogeneous deformations, indicating systematic mismatches in both size and shape. In contrast, the ANTsPy template showed mean scaling and deformation values near 1.0, reflecting closer correspondence to real human anatomy. These findings demonstrate that the MNI152 template does not accurately represent the morphology of the studied population and that linear registration alone cannot correct its inherent biases. Reliance on MNI152 may therefore introduce unintended distortions in applications requiring anatomically realistic head models. More robust, unbiased template-generation pipelines, and potentially demographic-specific templates, may be necessary for improved accuracy.

neuroscience↗

bioRxiv2026

Assessing suitability of ultra-low-field MRI for TES electric field models

ObjectiveRecent ultra-low-field (ULF) MRI systems provide an option for more accessible and cost-effective imaging to widely used higher field (HF, usually 3T) systems. However, it remains uncertain whether ULF scans can be used for reliable TES current-flow modeling as opposed to using 3T MRI. MethodsWe used raw anatomical scans of 23 healthy adults from 64 mT and 3T systems. We enhanced the 64 mT scans using SynthSR. Using our established current-flow modeling pipeline we determined the electric fields (EF) for the ULF and 3T scans induced by the classic M1-SO electrode montage. We quantified the differences in EF at three regions of interest (ROI): left motor cortex (LMC), anterior commissure (AC), and left hippocampus (LHC). ResultsAt the cortical ROI (LMC), the individual mean EF based on a 3T scan was found to be 27% higher than the corresponding ULF scan. The average mean relative error was found to be 40%. At AC, the individual mean EF based on a 3T scan continued to be elevated by about 13% in magnitude with respect to the corresponding ULF scan with a mean error range of 31-34%. At LHC, comparison of the induced individual mean EF magnitudes indicated no clear bias between 3T-based and ULF-based models. The average mean relative error at this level dropped to [~]20%. DiscussionThe substantial decrease in cortical EF when using an ULF scan is predominantly explained by the overestimation of cerebrospinal fluid (CSF) noted by prior morphological comparison studies. Scaling up EF magnitude to match predicted cortical values to a 3T scan may be considered depending on the modeling question being addressed. While preliminary, our results suggest an "equalizing" effect between ULF and 3T predictions with increasing depth within the brain.

bioengineering↗

bioRxiv2025

53BP1 condensates function as bioreactors for NHEJ directed DNA repair and insulators to determine pathway choice

Genomic integrity requires efficient resolution of DNA damage. Non-homologous end joining (NHEJ) is the primary mechanism of DNA double strand break (DSB) repair in mammalian cells and is mediated by 53BP1, a tumor suppressor involved in preventing DSB end-resection and homologous recombination. NHEJ repair foci form following DSB formation, however how mesoscale assembly occurs and whether 53BP1 is the driver of this process are unknown. Also, despite knowledge of the identify of key pathway molecules, the specific functions of mesoscale repair condensates in DNA repair and pathway selectivity are unknown. To address these gaps, we determined the minimal domain of 53BP1 sufficient for phase separation in vitro and identified the key residues that governing its condensation. Utilizing a separation-of-function mutant, we demonstrate that 53BP1and its protein condensation is the core driver of NHEJ foci formation. 53BP1 condensates function as bioreactor compartments, increasing the effective concentration of substrates near double strand breaks and are essential for efficient DNA damage resolution. Additionally, we show that 53BP1 condensates function as insulators around DSB sites to prevent end-resection and direct repair pathway selectivity. Collectively our work reveals a specialized compartment for DNA repair through the spatial clustering of 53BP1 molecules into repair foci essential to maintain genome integrity.

cell biology↗

bioRxiv2025

Scalable and Adaptive Spatiotemporal Modeling for Task-Based fMRI Analysis

Task-based fMRI is commonly analyzed using voxel-wise general linear models, a non-spatial scalable approach that can yield fragmented activation maps. Spatial alternatives such as kernel smoothing and Bayesian models address this but either blur activation boundaries or are computationally prohibitive at modern spatial resolutions. We introduce SPLASH (Spline-Based Processing for Localized Adaptive Spatial Hemodynamics), a spatially adaptive and scalable framework based on localized thin-plate spline regression within brain parcels. Its spatial flexibility allows SPLASH to adapt to heterogeneous cortical organization and to generalize across diverse spatial domains. Using its hierarchical structure, we introduce a two-stage selective inference procedure that ensures valid false discovery rate control at the parcel and voxel levels. In simulations, SPLASH consistently delivered the best overall performance: its MSE was typically only 20-40% of that of prior spatial models, and both FPR and FNR remained well controlled. SPLASH also remained stable across smoothing choices and required only 2% of the computation time of Bayesian spatial approaches. Applied to Human Connectome Project data, SPLASH produced sharper activation patterns consistent with the motor homunculus and demonstrated higher reproducibility. SPLASH provides a generalizable, spatially adaptive, and scalable framework that strengthens statistical inference and improves neuroscientific interpretability in large-scale fMRI studies.

neuroscience↗

bioRxiv2025

CNS Myelin Sheath Lengths Locally Scale to Axon Diameter via Piezo1

Myelin sheath lengths vary by an order of magnitude in the central nervous system (CNS) and tune the timing of neuronal signaling. Thus, variation in myelin sheath length has been proposed to coordinate the timing of neuronal signaling to ultimately impact behavior. The mechanisms to establish myelin sheath length are unknown. For decades, reports have documented that in vivo myelin sheath size scales with the diameter of the ensheathed axon. We previously demonstrated diameter is sufficient to instruct myelin sheath lengths formed by oligodendrocytes using a synthetic axon culture system. The mechanisms of oligodendrocyte diameter-sensing and its translation into sheath elongation are still unknown. Here we demonstrate that diameter-sensing and sheath length is locally regulated: each individual myelin sheath responds to the underlying fiber diameter. We uncover a novel mechanism for scaling myelin sheath length to fiber diameter, through mechanosensitive ion channel Piezo1. In vivo, Piezo1 impacts the elongation of myelin sheaths on large diameter axons, recapitulating our in vitro results. Yet, surprisingly, there is no impact on myelin thickness with conditional Piezo1 loss. We propose Piezo1 provides a mechanism to establish hard-wired myelin sheath patterns, where oligodendrocytes transduce axon diameter into generating myelin segments with vastly different lengths.

neuroscience↗

bioRxiv2025

SAM-DTI: A Spatial Attention Model for Drug-Target Interaction Prediction

Drug-target protein interaction (DTI) prediction is an important area of research in drug repurposing and discovery. Laboratory experiments for exploring DTI space need a lot of time and money due to the high volume of potential compounds. The past decades advancements in deep learning have made progress in efficient prediction of DTIs. This paper proposes a spatial attention model for predicting drug-target interactions (SAM-DTI) from labeled SMILES strings and protein sequences. SAM-DTI (i) tries to encapsulate the input features with more relevance to the biological binding of drug-target interactions, rather than models that solely extract features from individual inputs, and (ii) emphasizes giving dynamic importance to each pair of drug-target interactions. The model follows a knowledgeinspired sub-structure vocabulary to preprocess the inputs. It predicts accurate DTI values from the interactions between each SMILES sub-structure and protein substructure pair. SAM-DTI achieves competitive performance compared to state-of-the-art methods.

bioinformatics↗