Biology subjects

Kummerfeld, E.

Publications and source records attributed to Kummerfeld, E..

5 recordsLinked to original sources

bioRxiv2026

Locus-Level Transposable Element Profiling Resolves Division-Coupled Transcriptional Dynamics During Human Endoderm Specification

BackgroundTransposable elements (TEs) constitute nearly half of the human genome and are now recognized as significant contributors to mammalian gene regulatory networks. Despite this, most transcriptomic studies quantify TE expression at the subfamily level, which may obscure meaningful variation arising from individual insertion sites. Whether resolving TE expression to individual loci can reveal biologically distinct signals during stem cell differentiation has not been systematically characterized. ResultsWe re-analysed a published RNA-seq time course of FUCCI-h9 human embryonic stem cells differentiating into definitive endoderm (0-72 hour, seven time points, three division cycles, three biological replicates), quantifying expression in parallel at two complementary resolutions: TE subfamilies using TEtranscripts and individual TE loci using TElocal. The primary finding is that individual TE loci capture heterogeneous transcriptional responses at cell division boundaries that are entirely absent at the subfamily level. Within-division-state PC1 variance for TE loci was substantially elevated at the first division cycle (6.72; 95% bootstrap CI 0.63-8.48) compared with TE subfamilies (0.22; CI 0.04-0.34), with non-overlapping confidence intervals providing statistically robust support for the resolution advantage. Differential expression analysis identified over 18,000 dynamic TE loci across the time course, exceeding the 268 differentially expressed subfamilies, with alternating phases of silencing and reactivation resolved only at locus resolution. Differentially expressed TE loci were non-randomly enriched at superenhancers (fold enrichment: 9.1-23.5-fold; p<0.001 by permutation), with peak overlap at 36-48 hours coinciding with the second cell division and endoderm commitment. ERV1-class elements, particularly HERVH-int, were the dominant contributors, and representative loci near the endoderm regulators MIXL1 and ID3 showed differentiation-induced RNA-seq signal within proximal superenhancer domains. ConclusionsTE loci exhibit heterogeneous transcriptional responses at cell division boundaries, a signal with non-overlapping bootstrap confidence intervals relative to TE subfamilies at the first division cycle that is entirely masked by subfamily-level aggregation. This division-boundary resolution advantage, together with permutation-confirmed enrichment of dynamic loci at superenhancers during a discrete 36-48 hour endoderm commitment window, supports broader adoption of locus-resolved TE quantification as a complement to conventional gene expression analysis in developmental genomics.

bioinformatics↗

bioRxiv2024

Causal Discovery Analysis Reveals Insights into Psychosis Proneness, Brain Function, and Environmental Factors among Young Individuals

BackgroundExperiencing symptoms of psychosis, such as delusions and hallucinations, are observed in general, nonclinical populations. These experiences are sometimes described as psychosis proneness (PP) and potentially part of the psychosis continuum. The directional relationships among various factors contributing to psychosis proneness and its interactions, encompassing both environmental and neural mechanisms, lack comprehensive description. We aimed to identify targets to prevent psychosis proneness and its interactions by characterizing pathways using causal discovery analysis (CDA). MethodsParticipants were 194 healthy adolescent and young adult twin and sibling pairs aged between 14-24 years from Turkiye. They completed comprehensive assessments evaluating sociodemographic status, environmental risk, general intelligence, self-schema, PP, and working memory (WM) performance during fMRI (37 variables). CDA was applied, a novel machine learning algorithm, to understand the causal relationships of PP. ResultsThe analysis identified negative self-schema as having the largest causal effect among all assessments in PP [Effect size (ES)= 0.55]. Secondly, social cohesion and trust (SC&T) had a protective causal effect on PP [ES= -0.18]. Lastly, PP was identified as a direct cause of greater activation in DLPFC (BA9a-BA46v) during manipulation in the WM (ES= 0.14). ConclusionsCDA provides directionality of the 37 variables which were not presented earlier. The findings highlight the significance of negative self-schema and SC&T in the general population with PP, emphasizing the potential for preventive interventions targeting these factors. These findings also suggest a role for DLPFC as a potential target in this regard. To our knowledge, this is the first study using data-driven analysis to model causal mechanisms in PP in the general population.

neuroscience↗

bioRxiv2021

Executive and Attentional Resting-State Hubs of the Human Cortical Connectome Revealed by a Causal Discovery Method for Data-Driven Effective Connectivity Analysis

We demonstrate a data-driven approach for calculating a "causal connectome" of directed connectivity from resting-state fMRI data using a greedy adjacency search and pairwise non-Gaussian edge orientations. We used this approach to construct n=442 causal connectomes. These connectomes were very sparse in comparison to typical Pearson correlation-based graphs (roughly 2.25% edge density) yet were fully connected in nearly all cases. Prominent highly connected hubs of the causal connectome were situated in attentional (dorsal attention) and executive (frontoparietal and cingulo-opercular) networks. These hub networks had distinctly different connectivity profiles: attentional networks shared incoming connections with sensory regions and outgoing connections with higher cognitive networks, while executive networks primarily connected to other higher cognitive networks and had a high degree of bidirected connectivity. Virtual lesion analyses accentuated these findings, demonstrating that attentional and executive hub networks are points of critical vulnerability in the human causal connectome. These data highlight the central role of attention and executive control networks in the human cortical connectome and set the stage for future applications of data-driven causal connectivity analysis in psychiatry.

neuroscience↗

bioRxiv2021

Explainable Machine Learning Analysis Reveals Gender Differences in the Phenotypic and Neurobiological Markers of Cannabis Use Disorder

BackgroundCannabis Use Disorder (CUD) has been linked to environmental, personality, mental health, neurocognitive and neurobiological risk factors. While many studies have revealed gender differences in CUD, the relative importance of these complex factors by gender has not been described. MethodsWe conducted a data-driven examination of gender differences in CUD in a community sample of young adults (Human Connectome Project [HCP]; n = 1204, 54% female). We employed state-of-the-art machine learning methods [gradient tree boosting, XGBoost] in combination with novel factor ranking tools [SHapleys Additive exPlanations (SHAP)] as an explainable machine learning approach in the multimodal data collected by the HCP (phenotypic and brain data). ResultsWe were able to successfully classify both cannabis dependence and cannabis use levels. Previously identified environmental, personality, mental health, neurocognitive, and brain factors highly contributed to the classification. Predominantly-male risk factors included personality (high openness), mental health (high externalizing, high childhood conduct disorder, high fear somaticism), neurocognitive (impulsive delay discounting, slow working memory performance) and brain (low hippocampal volume) factors. Conversely, predominantly-female risk factors included environmental (low education level, low instrumental support) factors. ConclusionsOur data-driven analysis of gender differences in the multimodal risk factors underlying cannabis dependence and use levels demonstrate that environmental factors contribute more strongly to CUD in women, whereas individual factors such as personality, mental health and neurocognitive factors have a larger importance in men. This warrants further investigations, and suggests the importance of understanding how these differences relate to the development of effective treatment approaches.

neuroscience↗

bioRxiv2020

An Integrated Multimodal Model of Alcohol Use Disorder Generated by Data-Driven Causal Discovery Analysis

ObjectiveAlcohol use disorder (AUD) has high prevalence and adverse societal impacts, but our understanding of the factors driving AUD is hampered by a lack of studies that describe the complex multifactorial mechanisms driving AUD. MethodsWe used Causal Discovery Analysis (CDA) with data from the Human Connectome Project (HCP; n = 926 [54% female], 22% AUD [37% female]). Our outcome variable was number of AUD symptoms. We applied exploratory factor analysis (EFA) to parse phenotypic measures into underlying constructs, and assessed functional connectivity within 12 resting-state brain networks as an indicator of brain function. We then employed data-driven CDA to generate an integrated model relating phenotypic factors, fMRI network connectivity, and AUD symptom severity. ResultsEFA extracted 18 factors representing the wide HCP phenotypic space (100 measures). CDA produced an integrated multimodal model, highlighting a limited set of causes of AUD. The model proposed a hierarchy with causal influence propagating from brain function to cognition (fluid/crystalized cognition, language & working memory) to social (agreeableness/social support) to affective/psychiatric function (negative affect, low conscientiousness/attention, externalizing symptoms) and ultimately AUD severity. Every edge in the model was present at p < .001, and the SEM model overall provided a good fit (RMSEA = .06, Tucker-Lewis Index = .91). ConclusionsOur data-driven model confirmed hypothesized influences of cognitive and affective factors on AUD, while underscoring that traditional addiction models need to be expanded to highlight the importance of social factors, amongst others. Results further demonstrated that it is possible to extract a limited set of causal factors of AUD, which can inform future research aimed at tracking factors that dynamically predict alcohol use trajectories. Lastly, the presented model identified potential treatment targets for AUD, including neuromodulation of the frontoparietal network, cognitive/affective interventions, and social interventions.

neuroscience↗