Biology subjects

Thomas, R. M.

Publications and source records attributed to Thomas, R. M..

8 recordsLinked to original sources

bioRxiv2023

Harmonization techniques for machine learning studies using multi-site functional MRI data

In recent years, the collection and sharing of resting-state functional magnetic resonance imaging (fMRI) datasets across multiple centers have enabled studying psychiatric disorders at scale, and prompted the application of statistically powerful tools such as deep neural networks. Yet, multi-center datasets introduce non-biological heterogeneity that can confound the biological signal of interest and produce erroneous findings. To mitigate this problem, the neuroimaging community has adopted harmonization techniques previously proposed in other domains to remove site-effects from fMRI data. The reported success of these approaches in improving the generalization of the models have varied significantly. It remains unclear whether harmonization techniques could boost the final outcome of multi-site fMRI studies, to what extent, and which approaches are best suited for this task. In an attempt to objectively answer these questions, we conduct a standardized rigorous evaluation of seven different harmonization techniques from the neuroimaging and computer vision literature on two large-scale multi-site datasets (N = 2169 and N = 2366) to diagnose autism spectrum disorder and major depression disorder from static and dynamic representations of fMRI data. Interestingly, while all harmonization techniques removed site-effects from the data, they had little influence on disorder classification performance in standard k-fold and leave-one-site-out validation settings over a well-tuned baseline. Further investigation shows that the baseline model implicitly learns site-invariant features which could well explain its competitiveness with explicit harmonization techniques and suggest orthogonality between latent disease features and site discrminative features. However, additional experiments show that harmonization methods could be critical to report faithful results in settings where there is high intra-site class imbalance and the learning algorithm is prone to overfit on spurious features confounding the final outcome of the study.

neuroscience↗

bioRxiv2023

Foxp3 depends on Ikaros for control of regulatory T cell gene expression and function

Ikaros is a transcriptional factor required for conventional T cell development, differentiation, and anergy. While the related factors Helios and Eos have defined roles in regulatory T cells (Treg), a role for Ikaros has not been established. To determine the function of Ikaros in the Treg lineage, we generated mice with Treg-specific deletion of the Ikaros gene (Ikzf1). We find that Ikaros cooperates with Foxp3 to establish a major portion of the Treg epigenome and transcriptome. Ikaros-deficient Treg exhibit Th1-like gene expression with abnormal expression of IL-2, IFNg, TNFa, and factors involved in Wnt and Notch signaling. While Ikzf1-Treg-cko mice do not develop spontaneous autoimmunity, Ikaros-deficient Treg are unable to control conventional T cell-mediated immune pathology in response to TCR and inflammatory stimuli in models of IBD and organ transplantation. These studies establish Ikaros as a core factor required in Treg for tolerance and the control of inflammatory immune responses.

immunology↗

bioRxiv2023

Chromatin conformation dynamics during CD4+ T cell activation implicates autoimmune disease-associated genes and regulatory elements

Genome-wide association studies (GWAS) have identified hundreds of genetic signals associated with autoimmune disease. The majority of these signals are located in non-coding regions and likely impact cis-regulatory elements (cRE). Because cRE function is dynamic across cell types and states, profiling the epigenetic status of cRE across physiological processes is necessary to characterize the molecular mechanisms by which autoimmune variants contribute to disease risk. We localized risk variants from 15 autoimmune GWAS to cRE active during TCR-CD28 costimulation of naive human CD4+ T cells. To characterize how dynamic changes in gene expression correlate with cRE activity, we measured transcript levels, chromatin accessibility, and promoter-cRE contacts across three phases of naive CD4+ T cell activation using RNA-seq, ATAC-seq, and HiC. We identified [~]1,200 protein-coding genes physically connected to accessible disease-associated variants at 423 GWAS signals, at least one-third of which are dynamically regulated by activation. From these maps, we functionally validated a novel stretch of evolutionarily conserved intergenic enhancers whose activity is required for activation-induced IL2 gene expression in human and mouse, and is influenced by autoimmune-associated genetic variation. The set of genes implicated by this approach are enriched for genes controlling CD4+ T cell function and genes involved in human inborn errors of immunity, and we pharmacologically validated eight implicated genes as novel regulators of T cell activation. These studies directly show how autoimmune variants and the genes they regulate influence processes involved in CD4+ T cell proliferation and activation.

immunology↗

bioRxiv2023

Multimodal Data Integration Advances Longitudinal Prediction of the Naturalistic Course of Depression and Reveals a Multimodal Signature of Disease Chronicity

The ability to individually predict disease course of major depressive disorder (MDD) is essential for optimal treatment planning. Here, we use a data-driven machine learning approach to assess the predictive value of different sets of biological data (whole-blood proteomics, lipid-metabolomics, transcriptomics, genetics), both separately and added to clinical baseline variables, for the longitudinal prediction of 2-year MDD chronicity (defined as presence of MDD diagnosis after 2 years) at the individual subject level. Prediction models were trained and cross-validated in a sample of 643 patients with current MDD (2-year chronicity n = 318) and subsequently tested for performance in 161 MDD individuals (2-year chronicity n = 79). Proteomics data showed best unimodal data predictions (AUROC = 0.68). Adding proteomic to clinical data at baseline significantly improved 2-year MDD chronicity predictions (AUROC = 0.63 vs AUROC = 0.78, p = 0.013), while the addition of other -omics data to clinical data did not yield significantly increased model performance. SHAP and enrichment analysis revealed proteomic analytes involved in inflammatory response and lipid metabolism, with fibrinogen levels showing the highest variable importance, followed by symptom severity. Machine learning models outperformed psychiatrists ability to predict two-year chronicity (balanced accuracy = 71% vs 55%). This study showed the added predictive value of combining proteomic, but not other -omic data, with clinical data. Adding other -omic data to proteomics did not further improve predictions. Our results reveal a novel multimodal signature of MDD chronicity that shows clinical potential for individual MDD disease course predictions from baseline measurements.

genomics↗

bioRxiv2019

Human follicular helper T cell promoter connectomes reveal novel genes and regulatory elements at SLE GWAS loci

Systemic lupus erythematosus (SLE) is a complex inflammatory disease mediated by autoreactive antibodies that damages multiple tissues in children and adults. Genome-wide association studies (GWAS) have statistically implicated hundreds of loci in the susceptibility to human disease, including SLE, but the majority have failed to identify the causal variants or the effector genes. As a physicochemical approach to detecting functional variants and connecting them to target genes, we generated comprehensive, high-resolution maps of SLE variant accessibility and gene connectivity in the context of the three-dimensional chromosomal architecture of human tonsillar follicular helper T cells (TFH), a cell type required for the production of anti-nuclear antibodies characteristic of SLE. These spatial epigenomic maps identified a shortlist of over 400 potentially functional variants across 48 GWAS-implicated SLE loci. Twenty percent of these variants were located in open promoters of highly-expressed TFH genes, while 80% reside in non-promoter genomic regions that are connected in 3D to genes that likewise tend to be highly expressed in TFH. Importantly, we find that 90% of SLE-associated variants exhibit spatial proximity to genes that are not nearby in the 1D sequence of the genome, and over 60% of variants skip the nearest gene to physically interact only with the promoters of distant genes. Gene ontology confirmed that genes in spatial proximity to SLE variants reside in highly SLE-relevant networks, including accessible variants that loop 200-1000 kb to interact with the promoters of the canonical TFH genes BCL6 and CXCR5. CRISPR-Cas9 genome editing confirmed that these variants reside in novel, distal regulatory elements required for normal BCL6 and CXCR5 expression by T cells. Furthermore, SLE-associated SNP-promoter interactomes implicated a set of novel genes with no known role in TFH or SLE disease biology, including the homeobox-interacting protein kinase HIPK1 and the Ste kinase homolog MINK1. Targeting these kinases in primary human TFH cells inhibited production of IL-21, a requisite cytokine for production of class-switched antibodies by B cells. This 3D-variant-to-gene mapping approach gives mechanistic insight into the SLE-associated regulatory architecture of the human genome.

genomics↗

bioRxiv2019

Individual Prediction of Psychotherapy Outcome in Posttraumatic Stress Disorder using Neuroimaging Data

ObjectiveTrauma-focused psychotherapy is the first-line treatment for posttraumatic stress disorder (PTSD) but 30-50% of patients do not benefit sufficiently. We investigated whether structural and resting-state functional magnetic resonance imaging (MRI/rs-fMRI) data could distinguish between treatment responders and non-responders on the group and individual level.\n\nMethodsForty-four male veterans with PTSD underwent baseline scanning followed by trauma-focused psychotherapy. Voxel-wise gray matter volumes were extracted from the structural MRI data and resting-state networks (RSNs) were calculated from rs-fMRI data using independent component analysis. Data were used to detect differences between responders and non-responders on the group level using permutation testing, and the single-subject level using Gaussian process classification with cross-validation.\n\nResultsA RSN centered on the bilateral superior frontal gyrus differed between responders and non-responder groups (PFWE < 0.05) while a RSN centered on the pre-supplementary motor area distinguished between responders and non-responders on an individual-level with 81.4% accuracy (P < 0.001, 84.8% sensitivity, 78% specificity and AUC of 0.93). No significant single-subject classification or group differences were observed for gray matter volume.\n\nConclusionsThis proof-of-concept study demonstrates the feasibility of using rs-fMRI to develop neuroimaging biomarkers for treatment response, which could enable personalized treatment of patients with PTSD.

neuroscience↗

bioRxiv2018

Cingulate dependent social risk assessment in rats

Social transmission of distress has been conceived of as a one-way phenomenon in which an observer catches the emotions of another. Here we use a paradigm in which an observer rat witnesses another receive electro-shocks. Bayesian model comparison and Granger causality argue against this one-way vision in favor of bidirectional information transfer: how the observer reacts to the demonstrators distress influences the behavior of the demonstrator. Intriguingly, this was true to a similar extent across highly familiar and entirely unfamiliar rats. Injecting muscimol in the anterior cingulate of observers reduced freezing in the observers and in the demonstrators receiving the shocks. That rats share the distress of unfamiliar strains is at odds with evolutionary thinking that empathy should be biased towards close individuals. Using simulations, we support the complementary notion that distress transmission could be selected to more efficiently detect dangers in a group.

neuroscience↗

bioRxiv2018

Where and how our brain represents the temporal structure of observed action

Reacting faster to the behavior of others provides evolutionary advantages. Reacting to unpredictable events takes hundreds of milliseconds. Understanding where and how the brain represents what actions are likely to follow one another is therefore important. Everyday actions are predictable sequences of acts, yet neuroscientists focus on how brains responds to unexpected, individual motor acts. Using fMRI we show the brain encodes sequence-specific information in the motor system. Using EEG, we show visual responses were faster and smaller for predictable sequences that recruit the motor system. This study shifts the study of action observation from single acts to motor sequences, informs how we adapt to the actions of others and suggests the motor system may implement perceptual predictive coding.

neuroscience↗