Biology subjects

Gold, J.

Publications and source records attributed to Gold, J..

9 recordsLinked to original sources

bioRxiv2026

Multimodal spatial alignment and morphology mapping with MOSAICField

Recent efforts to build comprehensive tissue and tumor atlases leverage diverse spatial technologies to measure transcriptomic, proteomic, epigenetic, and other modalities with hundreds to thousands of features at thousands to millions of spatially resolved locations in a tissue slice. Integrating such data across spatial technologies that differ in molecular features, spatial resolution, and tissue morphology remains a major challenge. We introduce MultimOdal Spatial Alignment and Integration with Coordinate neural Field (MOSAICField), a unified framework for aligning spatial slices across arbitrary combinations of experimental modalities. MOSAICField computes two types of spatial alignments across multiple slices from the same tissue: physical alignment, which reconstructs a contiguous 3D model of the original tissue, and morphological alignment, which maps distinct morphological or anatomical structures, such as ducts, veins, or neurons, that may traverse the tissue at different angles relative to the direction of slicing. MOSAICField computes both alignments using a deep neural network that estimates a nonlinear deformation field with a multimodal feature loss. We evaluate MOSAICField on simulated data and a prostate cancer sample from the Human Tumor Atlas Network (HTAN), containing more than a dozen spatial slices with multimodal profiling data. MOSAICField constructs an accurate 3D tumor model, tracks the architecture of the prostatic ductal system, and improves analysis of features within and across modalities, outperforming existing methods.

bioinformatics↗

bioRxiv2025

Learning Latent Trajectories in Developmental Time Series with Hidden-Markov Optimal Transport

Deriving the sequence of transitions between cell types, or differentiation events, that occur during organismal development is one of the fundamental challenges in developmental biology. Single-cell and spatial sequencing of samples from different developmental timepoints provide data to investigate differentiation but inferring a sequence of differentiation events requires: (1) finding trajectories, or ancestor:descendant relationships, between cells from consecutive timepoints; (2) coarse-graining these trajectories into a differentiation map, or collection of transitions between cell types, rather than individual cells. We introduce Hidden-Markov Optimal Transport (HM -OT), an algorithm that simultaneously groups cells into cell types and learns transitions between these cell types from developmental transcriptomics time series. HM -OT uses low-rank optimal transport to simultaneously align samples in a time series and learn a sequence of clusterings and a differentiation map with minimal total transport cost. We assume that the law governing cell-type trajectories is characterized by the joint law on consecutive time points, tantamount to a Markov assumption on these latent trajectories. HM -OT can learn these clusterings in a fully unsupervised manner or can generate the least-cost cell type differentiation map consistent with a given set of cell type labels. We validate the unsupervised clusters and cell type differentiation map output by HM -OT on a Stereo-seq dataset of zebrafish development, and we demonstrate the scalability of HM -OT to a massive Stereo-seq dataset of mouse embryonic development. Code availabilitySoftware is available at https://github.com/raphael-group/HM-OT

bioinformatics↗

bioRxiv2024

DeST-OT: Alignment of Spatiotemporal Transcriptomics Data

Spatially resolved transcriptomics (SRT) measures mRNA transcripts at thousands of locations within a tissue slice, revealing spatial variations in gene expression and distribution of cell types. In recent studies, SRT has been applied to tissue slices from multiple timepoints during the development of an organism. Alignment of this spatiotemporal transcriptomics data can provide insights into the gene expression programs governing the growth and differentiation of cells over space and time. We introduce DeST-OT (Developmental SpatioTemporal Optimal Transport), a method to align SRT slices from pairs of developmental timepoints using the framework of optimal transport (OT). DeST-OT uses semi-relaxed optimal transport to precisely model cellular growth, death, and differentiation processes that are not well-modeled by existing alignment methods. We demonstrate the advantage of DeST-OT on simulated slices. We further introduce two metrics to quantify the plausibility of a spatiotemporal alignment: a growth distortion metric which quantifies the discrepancy between the inferred and the true cell type growth rates, and a migration metric which quantifies the distance traveled between ancestor and descendant cells. DeST-OT outperforms existing methods on these metrics in the alignment of spatiotemporal transcriptomics data from the development of axolotl brain. Code availabilitySoftware is available at https://github.com/raphael-group/DeST_OT

bioinformatics↗

bioRxiv2023

Heterogeneity in Functional Connectivity: Dimensional Predictors of Individual Variability during Rest and Task fMRI in Psychosis

BackgroundIndividuals with schizophrenia spectrum disorders (SSD) often demonstrate cognitive impairments, associated with poor functional outcomes. While neurobiological heterogeneity has posed challenges when examining social cognition in SSD, it provides a unique opportunity to explore brain-behavior relationships. We examined the relationship between behavioral data and individual variability of functional connectivity at rest and during an emotional-processing task. MethodsNeuroimaging and behavioral data were analyzed for 193 individuals with SSD and 155 controls (total n = 348). Individual variability was quantified through mean correlational distance (MCD) of functional connectivity between participants; MCD was defined as a global variability score. Hierarchical regressions were performed on variability scores derived from resting state and Empathic Accuracy (EA) task functional connectivity data to determine potential predictors (e.g., age, sex, neurocognitive and social cognitive scores) of individual variability. ResultsSSD showed greater MCD during rest (p = 0.00013) and task (p = 0.022). In the hierarchical regression, diagnosis remained significant when social cognition was included during rest (p = 0.008), but not during task (p = 0.50); social cognition was significant during both rest and task (both p = 0.01). ConclusionsDiagnostic differences were more prevalent during unconstrained resting scans, whereas the task pushed participants into a more common pattern which better emphasized transdiagnostic differences in cognitive abilities. Focusing on variability may provide new opportunities for interventions targeting specific cognitive impairments to improve functional outcomes.

neuroscience↗

bioRxiv2020

White matter brain aging In Relationship to Schizophrenia and Its Cognitive Deficit

We hypothesized that cerebral white matter deficits in schizophrenia (SZ) are driven in part by accelerated white matter aging and are associated with cognitive deficits. We used machine learning model to predict individual age from diffusion tensor imaging features and calculated the delta age ({Delta}age) as the difference between predicted and chronological age. Through this approach, we translated multivariate white matter imaging features into an age-scaled metric and used it to test the temporal trends of accelerated aging-related white matter deficit in SZ and its association with the cognition. Followed feature selection, a machine learning model was trained with fractional anisotropy values in 34 of 43 tracts on a training set consisted of 107 healthy controls (HC). The brain age of 166 SZs and 107 HCs in the testing set were calculated using this model. Then, we examined the SZ-HC group effect on {Delta}age and whether this effect was moderated by chronological age using the regression spline model. The results showed that {Delta}age was significantly elevated in the age >30 group in patients (p < 0.001) but not in age [<=] 30 group (p = 0.364). {Delta}age in patients was significantly and negatively associated with both working memory ({beta} = -0.176, p = 0.007) and processing speed ({beta} = -0.519, p = 0.035) while adjusting sex and chronological age. Overall, these findings indicate that the {Delta}age is elevated in SZs and become significantly from middle life stage; the increase of {Delta}age in SZs is associated with the decline neurocognitive performance.

neuroscience↗

bioRxiv2019

Dynamic Reorganization of the Frontal Parietal Network During Cognitive Control and Episodic Memory

Higher cognitive functioning is supported by adaptive reconfiguration of large-scale functional brain networks. Cognitive control (CC), which plays a vital role in flexibly guiding cognition and behavior in accordance with our goals, supports a range of executive functions via distributed brain networks. These networks process information dynamically and can be represented as functional connectivity changes between network elements.\n\nUsing graph theory, we explored context-dependent network reorganization in 56 healthy adults performing fMRI tasks from two cognitive domains that varied in CC and episodic-memory demands. We examined whole-brain modular structure during the DPX task, which engages proactive CC in the frontal-parietal cognitive control network (FPN), and the RiSE task, which manipulates CC demands at encoding and retrieval during episodic-memory processing, and engages FPN, the medial-temporal lobe and other memory related networks in a context dependent manner.\n\nAnalyses revealed different levels of network integration and segregation. High CC conditions exhibited greater integration across tasks. Moreover, nodes associated with higher cognitive functioning displayed the greatest amount of dynamic module reorganization across tasks. Specifically, the FPN displayed a high level of segregation in the DPX task, where only demands for proactive control varied, and more complex network integration with default mode and salience networks in the RiSE task, where CC is differentially integrated during memory encoding and retrieval. These findings provide insight into how brain networks reorganize to support differing task contexts, suggesting that the FPN flexibly segregates during focused proactive control and integrates to support control in other domains such as episodic-memory.

neuroscience↗

bioRxiv2019

Post-decision processing in primate prefrontal cortex influences subsequent choices on an auditory decision-making task

Perceptual decisions do not occur in isolation but instead reflect ongoing evaluation and adjustment processes that can affect future decisions. However, the neuronal substrates of these across-decision processes are not well understood, particularly for auditory decisions. We measured and manipulated the activity of choice-selective neurons in the ventrolateral prefrontal cortex (vlPFC) while monkeys made decisions about the frequency content of noisy auditory stimuli. As the decision was being formed, vlPFC activity was not modulated strongly by the task. However, after decision commitment, vlPFC population activity encoded the sensory evidence, choice, and outcome of the current trial and predicted subject-specific choice biases on the subsequent trial. Consistent with these patterns of neuronal activity, electrical microstimulation in vlPFC tended to affect the subsequent, but not current, decision. Thus, distributed post-commitment representations of graded decision-related information in prefrontal cortex can play a causal role in evaluating past decisions and biasing subsequent ones.

neuroscience↗

bioRxiv2018

The Valuation of Social Rewards in Schizophrenia

Social impairment in schizophrenia is often thought to reflect poor social cognition. Here we examine responses to social rewards, an aspect of social functioning that is not featured prominently in the literature. The goal of this experiment was to explore whether people with schizophrenia (1) undervalue social rewards, and (2) whether the undervaluation of social rewards was related to social motivation and pleasure deficits in schizophrenia and decreased social functioning. People with schizophrenia and healthy participants completed a game (Shore & Heerey, 2011) to explore preferences for different types of social (polite and genuine smiles) and nonsocial (monetary) rewards from computerized opponents. Preferences for reward types were quantified for each participant based on choice behavior during the game. Participants also completed a smile discrimination task to assess their ability to discriminate these types of smiles. Analyses revealed that people with schizophrenia (N = 41) treated genuine smiles as significantly less rewarding than did healthy participants (N = 29), despite showing a similar preference for monetary rewards. Interestingly, the undervaluation of social rewards was not related to the ability to discriminate between the smiles. The current findings provide preliminary evidence of reduced social reward valuation among individuals with schizophrenia, which may have implications for behavior in face-to-face social interactions.\n\nGeneral Scientific SummarySocial cognitive deficits are well documented in schizophrenia. However, it is less clear how people with schizophrenia respond to social rewards. Our study yields initial evidence of a deficit in social reward valuation (operationalized as a preference for genuine smiles) despite similar preferences for monetary rewards.

neuroscience↗

bioRxiv2018

Rapid Whole Genome Sequencing Decreases Morbidity and Healthcare Cost of Hospitalized Infants

BACKGROUNDGenetic disorders are a leading cause of morbidity and mortality in infants. Rapid Whole Genome Sequencing (rWGS) can diagnose genetic disorders in time to change acute medical or surgical management (clinical utility) and improve outcomes in acutely ill infants.\n\nMETHODSRetrospective cohort study of acutely ill inpatient infants in a regional childrens hospital from July 2016-March 2017. Forty-two families received rWGS for etiologic diagnosis of genetic disorders. Probands received standard genetic testing as clinically indicated. Primary end-points were rate of diagnosis, clinical utility, and healthcare utilization. The latter was modelled in six infants by comparing actual utilization with matched historical controls and/or counterfactual utilization had rWGS been performed at different time points.\n\nFINDINGSThe diagnostic sensitivity was 43% (eighteen of 42 infants) for rWGS and 10% (four of 42 infants) for standard of care (P=.0005). The rate of clinical utility for rWGS (31%, thirteen of 42 infants) was significantly greater than for standard of care (2%, one of 42; P=.0015). Eleven (26%) infants with diagnostic rWGS avoided morbidity, one had 43% reduction in likelihood of mortality, and one started palliative care. In six of the eleven infants, the changes in management reduced inpatient cost by $800, 000 to $2,000,000.\n\nDISCUSSIONThese findings replicate a prior study of the clinical utility of rWGS in acutely ill inpatient infants, and demonstrate improved outcomes and net healthcare savings. rWGS merits consideration as a first tier test in this setting.

clinical trials↗