Kai Ueltzhöffer

Congenital heart disease (CHD) originates during early embryogenesis, yet when disease genes act cannot be determined from existing data. Human fetal heart atlases lack critical time points, and perturbation screens that systematically assess gene function currently profile at most a few time points. Here we combined single-cell CRISPR screening of 400 disease-relevant genes in cardioids, sampled daily over seven developmental days (~975,000 cells), with a matched multiome atlas. Perturbation phenotypes were strongly dependent on developmental stage. Perturbations that reshaped cell-fate allocation or gene expression were enriched for CHD-linked genes, and most acted already during progenitor stages. Timing was largely independent of the target gene's own expression, marking instead when a cell state became functionally dependent on it, and some effects persisted after expression declined. Consensus gene regulatory networks integrating perturbation and multiome data recovered stage-specific regulators and nominated HOXB2 and HOXB3 as candidate cardiac regulators. Many perturbations converged on shared downstream programmes, which were reorganised through differentiation. Effects correlated in progenitors were maintained in differentiated cells, split into distinct programmes, converged with other perturbations or disappeared entirely, indicating that gene function is highly context dependent. Beyond advancing our understanding of the developmental regulatory logic underlying CHD, our work provides a resource for developing context-dependent models for in silico prediction of perturbation effects.

Protected moral values facilitate empathic concern for others, who are exposed to an existential threat, so that one spontaneously helps without taking into account utilitarian cost–benefit considerations. Subjects scoring high on the “Dark Triad” machiavellism, psychopathy, and narcissism are prone to ignore such appeals for selfless help. Until now, data on moral processing and moral decision-making following requests for altruistic help, which directly contrast appeals to protected and non-protected values in subjects with high and low scores on Dark Triad traits, have been missing. In this pilot study 25 healthy subjects with high and 27 with low Dark Triad scores participated in this functional magnetic resonance imaging study. We used a script-driven imagery paradigm to directly contrast requests for selfless help appealing to protected versus non-protected, negotiable moral values. Appeals to protected versus non-protected moral values elicited stronger activations in a large network including insula, amygdala, supramarginal gyrus, and dorsolateral prefrontal cortex. Non-protected values evoked stronger activation in superior frontal sulcus, occipito-temporal junction, and posterior cingulate cortex. During decision-making, high-scorers on the Dark Triad showed increased activations in the superior parietal lobule, precuneus, and intraparietal sulcus. Behaviorally, protected versus non-protected values strongly reduced the reliance on personal cost–benefit calculations in low-scorers, while high-scorers continued to rely on utilitarian deliberations. Data suggest that appeals to protected versus non-protected values activate distinct brain regions associated with strong moral emotions, other-directed cognition, and rule-based decision-making processes. High-scorers display an increased reliance on cost–benefit calculations, which persists even when protected values are threatened.

Inappropriate aggression is a prominent and clinically relevant interpersonal dysfunction of individuals with borderline personality disorder (BPD). Previous studies have shown that individuals with BPD interpret interpersonal signals in a hostile manner, but it is uncertain how this negativity bias impacts decision-making during aggressive encounters. In the present neuroimaging study, 48 medication-free women with BPD and 28 age- and intelligence-matched women played the Social Threat Aggression Paradigm (STAP), a competitive reaction time task in which the winner delivers an aversive sound blast to the loser. Crucially, in the STAP the alleged opponent displays either an angry or neutral facial expression at the beginning of each trial and selects increasingly loud blasts in order to provoke participants. Relative to healthy controls, women with BPD differentiated less between angry and neutral facial expressions, both in terms of aggressive behavior and of activity in medial prefrontal cortex, amygdala, and temporal pole. On the one hand, and contrary to our hypotheses, neural and behavioral responses to angry faces were reduced in women with BPD compared to healthy women. On the other hand, provocation heightened subsequent amygdala responses to neutral faces in BPD, and this was in turn associated with aggressive behavior, supporting a default negativity bias in BPD. The neurocognitive processes by which these alterations might guide aggressive behavior irrespective of interpersonal cues are presented and discussed. (PsycInfo Database Record (c) 2022 APA, all rights reserved).

In this treatment of random dynamical systems, we consider the existence—and identification—of conditional independencies at nonequilibrium steady-state. These independencies underwrite a particular partition of states, in which internal states are statistically secluded from external states by blanket states. The existence of such partitions has interesting implications for the information geometry of internal states. In brief, this geometry can be read as a physics of sentience, where internal states look as if they are inferring external states. However, the existence of such partitions—and the functional form of the underlying densities—have yet to be established. Here, using the Lorenz system as the basis of stochastic chaos, we leverage the Helmholtz decomposition—and polynomial expansions—to parameterise the steady-state density in terms of surprisal or self-information. We then show how Markov blankets can be identified—using the accompanying Hessian—to characterise the coupling between internal and external states in terms of a generalised synchrony or synchronisation of chaos. We conclude by suggesting that this kind of synchronisation may provide a mathematical basis for an elemental form of (autonomous or active) sentience in biology.

Dissipative accounts of structure formation show that the self-organisation of complex structures is thermodynamically favoured, whenever these structures dissipate free energy that could not be accessed otherwise. These structures therefore open transition channels for the state of the universe to move from a frustrated, metastable state to another metastable state of higher entropy. However, these accounts apply as well to relatively simple, dissipative systems, such as convection cells, hurricanes, candle flames, lightning strikes, or mechanical cracks, as they do to complex biological systems. Conversely, interesting computational properties—that characterize complex biological systems, such as efficient, predictive representations of environmental dynamics—can be linked to the thermodynamic efficiency of underlying physical processes. However, the potential mechanisms that underwrite the selection of dissipative structures with thermodynamically efficient subprocesses is not completely understood. We address these mechanisms by explaining how bifurcation-based, work-harvesting processes—required to sustain complex dissipative structures—might be driven towards thermodynamic efficiency. We first demonstrate a simple mechanism that leads to self-selection of efficient dissipative structures in a stochastic chemical reaction network, when the dissipated driving chemical potential difference is decreased. We then discuss how such a drive can emerge naturally in a hierarchy of self-similar dissipative structures, each feeding on the dissipative structures of a previous level, when moving away from the initial, driving disequilibrium.

Modern generative deep learning (DL) architectures allow for unsupervised learning of latent representations that can be exploited in several downstream tasks. Within the field of oncological medical imaging, we term these latent representations “digital tumor signatures” and hypothesize that they can be used, in analogy to radiomics features, to differentiate between lesions and normal liver tissue. Moreover, we conjecture that they can be used for the generation of synthetic data, specifically for the artificial insertion and removal of liver tumor lesions at user-defined spatial locations in CT images. Our approach utilizes an implicit autoencoder, an unsupervised model architecture that combines an autoencoder and two generative adversarial network (GAN)-like components. The model was trained on liver patches from 25 or 57 inhouse abdominal CT scans, depending on the experiment, demonstrating that only minimal data is required for synthetic image generation. The model was evaluated on a publicly available data set of 131 scans. We show that a PCA embedding of the latent representation captures the structure of the data, providing the foundation for the targeted insertion and removal of tumor lesions. To assess the quality of the synthetic images, we conducted two experiments with five radiologists. For experiment 1, only one rater and the ensemble-rater were marginally above the chance level in distinguishing real from synthetic data. For the second experiment, no rater was above the chance level. To illustrate that the “digital signatures” can also be used to differentiate lesion from normal tissue, we employed several machine learning methods. The best performing method, a LinearSVM, obtained 95% (97%) accuracy, 94% (95%) sensitivity, and 97% (99%) specificity, depending on if all data or only normal appearing patches were used for training of the implicit autoencoder. Overall, we demonstrate that the proposed unsupervised learning paradigm can be utilized for the removal and insertion of liver lesions at user defined spatial locations and that the digital signatures can be used to discriminate between lesions and normal liver tissue in abdominal CT scans.

This work combines the free energy principle and the ensuing active inference dynamics with recent advances in variational inference in deep generative models, and evolution strategies to introduce the “deep active inference” agent. This agent minimises a variational free energy bound on the average surprise of its sensations, which is motivated by a homeostatic argument. It does so by optimising the parameters of a generative latent variable model of its sensory inputs, together with a variational density approximating the posterior distribution over the latent variables, given its observations, and by acting on its environment to actively sample input that is likely under this generative model. The internal dynamics of the agent are implemented using deep and recurrent neural networks, as used in machine learning, making the deep active inference agent a scalable and very flexible class of active inference agent. Using the mountain car problem, we show how goal-directed behaviour can be implemented by defining appropriate priors on the latent states in the agent’s model. Furthermore, we show that the deep active inference agent can learn a generative model of the environment, which can be sampled from to understand the agent’s beliefs about the environment and its interaction therewith.