Insight
is our reward

Publications in Attention by NOMIS researchers

NOMIS Researcher(s)

Published in

August 4, 2023

Three-dimensional (3D) genomics shows immense promise for studying X chromosome inactivation (XCI) by interrogating changes to the X chromosomes’ 3D states. Here, we sought to characterize the 3D state of the X chromosome in naïve and primed human pluripotent stem cells (hPSCs). Using chromatin tracing, we analyzed X chromosome folding conformations in these cells with megabase genomic resolution. X chromosomes in female naïve hPSCs exhibit folding conformations similar to the active X chromosome (Xa) and the inactive X chromosome (Xi) in somatic cells. However, naïve X chromosomes do not exhibit the chromatin compaction typically associated with these somatic X chromosome states. In H7 naïve human embryonic stem cells, XIST accumulation observed on damaged X chromosomes demonstrates the potential for naïve hPSCs to activate XCI-related mechanisms. Overall, our findings provide insight into the X chromosome status of naïve hPSCs with a single-chromosome resolution and are critical in understanding the unique epigenetic regulation in early embryonic cells. Copyright © 2023 The Authors, some rights reserved.

Research field(s)
Health Sciences

NOMIS Researcher(s)

July 31, 2023

Supervised learning typically focuses on learning transferable representations from training examples annotated by humans. While rich annotations (like soft labels) carry more information than sparse annotations (like hard labels), they are also more expensive to collect. For example, while hard labels only provide information about the closest class an object belongs to (e.g., “this is a dog”), soft labels provide information about the object’s relationship with multiple classes (e.g., “this is most likely a dog, but it could also be a wolf or a coyote”). We use information theory to compare how a number of commonly-used supervision signals contribute to representation-learning performance, as well as how their capacity is affected by factors such as the number of labels, classes, dimensions, and noise. Our framework provides theoretical justification for using hard labels in the big-data regime, but richer supervision signals for few-shot learning and out-of-distribution generalization. We validate these results empirically in a series of experiments with over 1 million crowdsourced image annotations and conduct a cost-benefit analysis to establish a tradeoff curve that enables users to optimize the cost of supervising representation learning on their own datasets. © UAI 2023. All rights reserved.

Research field(s)
Health Sciences, Biomedical Research, Developmental Biology

NOMIS Researcher(s)

Published in

June 27, 2023

The human embryo undergoes morphogenetic transformations following implantation into the uterus, but our knowledge of this crucial stage is limited by the inability to observe the embryo in vivo. Models of the embryo derived from stem cells are important tools for interrogating developmental events and tissue–tissue crosstalk during these stages 1. Here we establish a model of the human post-implantation embryo, a human embryoid, comprising embryonic and extraembryonic tissues. We combine two types of extraembryonic-like cell generated by overexpression of transcription factors with wild-type embryonic stem cells and promote their self-organization into structures that mimic several aspects of the post-implantation human embryo. These self-organized aggregates contain a pluripotent epiblast-like domain surrounded by extraembryonic-like tissues. Our functional studies demonstrate that the epiblast-like domain robustly differentiates into amnion, extraembryonic mesenchyme and primordial germ cell-like cells in response to bone morphogenetic protein cues. In addition, we identify an inhibitory role for SOX17 in the specification of anterior hypoblast-like cells 2. Modulation of the subpopulations in the hypoblast-like compartment demonstrates that extraembryonic-like cells influence epiblast-like domain differentiation, highlighting functional tissue–tissue crosstalk. In conclusion, we present a modular, tractable, integrated 3 model of the human embryo that will enable us to probe key questions of human post-implantation development, a critical window during which substantial numbers of pregnancies fail. © 2023, The Author(s).

Research field(s)
Health Sciences, Psychology & Cognitive Sciences, Experimental Psychology

NOMIS Researcher(s)

Published in

May 2, 2023

Do we judge hate incidents similarly when they are performed using words or bodily actions? Hate speech incidents are rarely reported by bystanders, and whether or how much they should be punished remains a matter of legal, theoretical and social disagreement. In a pre-registered study (N = 1309), participants read about verbal and nonverbal attacks stemming from identical hateful intent, which created the same consequences for the victims. We asked them how much punishment the perpetrator should receive, how likely they would be to denounce such an incident and how much harm they judged the victim suffered. The results contradicted our pre-registered hypotheses and the predictions of dual moral theories, which hold that intention and harmful consequences are the sole psychological determinants of punishment. Instead, participants consistently rated verbal hate attacks as more deserving of punishment, denunciation and being more harmful to the victim than nonverbal attacks. This difference is explained by the concept of action aversion, suggesting that lay observers have different intrinsic associations with interactions involving words compared to bodily actions, regardless of consequences. This explanation has implications for social psychology, moral theories, and legislative efforts to sanction hate speech, which are considered. Protocol registration: The Stage 1 protocol for this Registered Report was accepted in principle on 29/06/2022. The protocol, as accepted by the journal, can be found at: https://doi.org/10.17605/OSF.IO/Z86TV . © 2023, The Author(s).

Research field(s)
Health Sciences, Psychology & Cognitive Sciences, Experimental Psychology

NOMIS Researcher(s)

Published in

January 19, 2023

Learning has been associated with modifications of synaptic and circuit properties, but the precise changes storing information in mammals have remained largely unclear. We combined genetically targeted voltage imaging with targeted optogenetic activation and silencing of pre- and post-synaptic neurons to study the mechanisms underlying hippocampal behavioral timescale plasticity. In mice navigating a virtual-reality environment, targeted optogenetic activation of individual CA1 cells at specific places induced stable representations of these places in the targeted cells. Optical elicitation, recording, and modulation of synaptic transmission in behaving mice revealed that activity in presynaptic CA2/3 cells was required for the induction of plasticity in CA1 and, furthermore, that during induction of these place fields in single CA1 cells, synaptic input from CA2/3 onto these same cells was potentiated. These results reveal synaptic implementation of hippocampal behavioral timescale plasticity and define a methodology to resolve synaptic plasticity during learning and memory in behaving mammals. © 2022 The Author(s)

Research field(s)
Health Sciences, Biomedical Research, Developmental Biology