Explore the vast range of titles available.

Animal models are used to understand principles of human biology. Within cognitive neuroscience, non-human primates are considered the premier model for studying decision-making behaviors in which direct manipulation experiments are still possible. Some prominent studies have brought to light major discrepancies between monkey and human cognition, highlighting problems with unverified extrapolation from monkey to human. Here, we use a parallel model system—artificial neural networks (ANNs)—to investigate a well-established discrepancy identified between monkeys and humans with a working memory task, in which monkeys appear to use a recency-based strategy while humans use a target-selective strategy. We find that ANNs trained on the same task exhibit a progression of behavior from random behavior (untrained) to recency-like behavior (partially trained) and finally to selective behavior (further trained), suggesting monkeys and humans may occupy different points in the same overall learning progression. Surprisingly, what appears to be recency-like behavior in the ANN, is in fact an emergent non-recency-based property of the organization of the neural network’s state space during its development through training. We find that explicit encouragement of recency behavior during training has a dual effect, not only causing an accentuated recency-like behavior, but also speeding up the learning process altogether, resulting in an efficient shaping mechanism to achieve the optimal strategy. Our results suggest a new explanation for the discrepency observed between monkeys and humans and reveal that what can appear to be a recency-based strategy in some cases may not be recency at all.

Amit Majithia and colleagues employ a pooled assay in human macrophages to assess the functional effects of all possible missense variants in PPARG . Their study shows the value of saturation mutagenesis and prospective experimental characterization to support diagnostic interpretation of newly discovered missense variants in disease-related genes. Clinical exome sequencing routinely identifies missense variants in disease-related genes, but functional characterization is rarely undertaken, leading to diagnostic uncertainty 1 , 2 . For example, mutations in PPARG cause Mendelian lipodystrophy 3 , 4 and increase risk of type 2 diabetes (T2D) 5 . Although approximately 1 in 500 people harbor missense variants in PPARG , most are of unknown consequence. To prospectively characterize PPARγ variants, we used highly parallel oligonucleotide synthesis to construct a library encoding all 9,595 possible single–amino acid substitutions. We developed a pooled functional assay in human macrophages, experimentally evaluated all protein variants, and used the experimental data to train a variant classifier by supervised machine learning. When applied to 55 new missense variants identified in population-based and clinical sequencing, the classifier annotated 6 variants as pathogenic; these were subsequently validated by single-variant assays. Saturation mutagenesis and prospective experimental characterization can support immediate diagnostic interpretation of newly discovered missense variants in disease-related genes.

Duplication or deficiency of the X‐linked MECP2 gene reliably produces profound neurodevelopmental impairment. MECP2 mutations are almost universally responsible for Rett syndrome (RTT), and particular mutations and cellular mosaicism of MECP2 may underlie the spectrum of RTT symptomatic severity. No clinically approved treatments for RTT are currently available, but human pluripotent stem cell technology offers a platform to identify neuropathology and test candidate therapeutics. Using a strategic series of increasingly complex human stem cell‐derived technologies, including human neurons, MECP2 ‐mosaic neurospheres to model RTT female brain mosaicism, and cortical organoids, we identified synaptic dysregulation downstream from knockout of MECP2 and screened select pharmacological compounds for their ability to treat this dysfunction. Two lead compounds, Nefiracetam and PHA 543613, specifically reversed MECP2‐ knockout cytologic neuropathology. The capacity of these compounds to reverse neuropathologic phenotypes and networks in human models supports clinical studies for neurodevelopmental disorders in which MeCP2 deficiency is the predominant etiology. Synopsis Deficiency of the X‐linked MECP2 gene profoundly impairs neurodevelopment. Clinically, mutations in MECP2 most commonly present as the severe and untreatable disease Rett syndrome. Innovative human pluripotent stem cell (PSC) technology enables the investigation of therapeutic candidates. Neurons differentiated from human MECP2 ‐KO PSCs showed altered expression of synapse‐relevant genes, synaptic morphology, and decreased calcium and network activities compared to control neurons. An in silico neural network simulation affirmed that synaptic structural parameters link with neural network activity, supporting our approach of rescuing synaptic structure and increasing activity. Strategic screening of select drug candidates with mechanisms of action that counteract MECP2 ‐KO deficiencies isolated two lead compounds, Nefiracetam and PHA 543613, for subsequent validation in 3D human cell models. MECP2 ‐mosaic neurospheres, a novel model of Rett female brain mosaicism, showed altered calcium activity that could be reversed by Nefiracetam and/or PHA 543613. Nefiracetam and/or PHA 543613 increased synaptic gene expression and reversed network pathology in human MECP2 ‐KO cortical organoids, arguing for their clinical trial in patients with MeCP2‐deficient neurodevelopmental disorders. Graphical Abstract Deficiency of the X‐linked MECP2 gene profoundly impairs neurodevelopment. Clinically, mutations in MECP2 most commonly present as the severe and untreatable disease Rett syndrome. Innovative human pluripotent stem cell (PSC) technology enables the investigation of therapeutic candidates.

Gene regulatory networks (GRNs) provide insights into the mechanisms of differential gene expression at a systems level. GRNs that relate to metazoan development have been studied extensively. However, little is still known about the design principles, organization and functionality of GRNs that control physiological processes such as metabolism, homeostasis and responses to environmental cues. In this study, we report the first experimentally mapped metazoan GRN of Caenorhabditis elegans metabolic genes. This network is enriched for nuclear hormone receptors (NHRs). The NHR family has greatly expanded in nematodes: humans have 48 NHRs, but C. elegans has 284, most of which are uncharacterized. We find that the C. elegans metabolic GRN is highly modular and that two GRN modules predominantly consist of NHRs. Network modularity has been proposed to facilitate a rapid response to different cues. As NHRs are metabolic sensors that are poised to respond to ligands, this suggests that C. elegans GRNs evolved to enable rapid and adaptive responses to different cues by a concurrence of NHR family expansion and modular GRN wiring. Synopsis Physical and/or regulatory interactions between transcription factors (TFs) and their target genes are essential to establish body plans of multicellular organisms during development, and these interactions have been studied extensively in the context of GRNs. The precise control of differential gene expression is also of critical importance to maintain physiological homeostasis, and many metabolic disorders such as obesity and diabetes coincide with substantial changes in gene expression. Much work has focused on the GRNs that control metazoan development; however, the design principles and organization of the GRNs that control systems physiology remain largely unexplored. In this study, we present the first gene‐centered GRN that includes ∼70 genes involved in C. elegans metabolism and physiology, 100 TFs and more than 500 protein–DNA interactions between them. The resulting metabolic GRN is enriched for NHRs, compared with other gene‐centered regulatory networks. NHRs are well‐known regulators of lipid meta‐qj;bolism in mammals. The transcriptional activity of NHRs can be modified by diffusible ligands, which allows these TFs to function as molecular sensors and rapidly alter the expression of their target genes. Interestingly, NHRs comprise the largest family of TFs in nematodes; the C. elegans genome encodes 284 NHRs, most of which are uncharacterized. Furthermore, their organization in GRNs has not yet been investigated. In our study, we show that the C. elegans NHRs that we retrieved in the metabolic GRN organize into network modules, and that most of these NHRs function to maintain lipid homeostasis in the nematode. Interestingly, network modularity has been proposed to facilitate rapid and robust changes in gene expression. Our results suggest that the C. elegans metabolic GRN may have evolved by combining NHR family expansion with the specific modular wiring of NHRs to enable the rapid adaptation of the animal to different environmental cues. NHRs can interact with transcriptional cofactors such as chromatin remodeling complexes and Mediator components. For instance, the C. elegans Mediator subunit, MDT‐15, can interact with NHR‐49 to regulate the expression of its target genes. To find all the TFs that MDT‐15 can interact with, we performed systematic yeast two‐hybrid assays with MDT‐15 versus 755 full‐length TFs. We found that MDT‐15 preferentially associates with NHRs, and specifically with those NHRs that confer a metabolic phenotype and that occur in the metabolic GRN. This illustrates the central role of MDT‐15 in the regulation of metabolic gene expression. Using a variety of genetic and biochemical approaches, we characterized NHR‐86 in more detail. NHR‐86 participates in one of the two NHR modules, and has a high‐flux capacity; that is it has both a high incoming and a high outgoing degree. We obtained an nhr‐86 mutant and generated an NHR‐86 antibody, and showed that NHR‐86 functions as an auto‐repressor in vivo and that nhr‐86 mutant animals store abnormally high levels of body fat. Finally, we discovered a novel NHR circuit that responds to nutrient availability. In this circuit NHR‐45 regulates the activity of nhr‐178 promoter in two distinct physiologically important tissues: the intestine and the hypodermis. Both of these NHRs are required to maintain lipid homeostasis in C. elegans . The expression of nhr‐178 is responsive to the nutritional status of the animal, which switches between ON and OFF states in the hypodermis. We found that NHR‐45 activity is necessary to control this switch in the hypodermis. Interestingly, NHR‐45 has opposite effects on the activity of the nhr‐178 promoter in these tissues: NHR‐45 activates this promoter in the intestine, but represses it in the hypodermis. Altogether our study leads to a model in which the expansion of the NHR family, TFs that have the capacity to act as fast molecular sensors, is combined with a modular network organization to enable rapid and robust responses to various environmental cues. We present the first gene regulatory network (GRN) that pertains to post‐developmental gene expression. Specifically, we mapped a transcription regulatory network of Caenorhabditis elegans metabolic gene promoters using gene‐centered yeast one‐hybrid assays. We found that the metabolic GRN is enriched for nuclear hormone receptors (NHRs) compared with other gene‐centered regulatory networks, and that these NHRs organize into functional network modules. The NHR family has greatly expanded in nematodes; C. elegans has 284 NHRs, whereas humans have only 48. We show that the NHRs in the metabolic GRN have metabolic phenotypes, suggesting that they do not simply function redundantly. The mediator subunit MDT‐15 preferentially interacts with NHRs that occur in the metabolic GRN. We describe an NHR circuit that responds to nutrient availability and propose a model for the evolution and organization of NHRs in C. elegans metabolic regulatory networks.

The prefrontal cortex encodes and stores numerous, often disparate, schemas and flexibly switches between them. Recent research on artificial neural networks trained by reinforcement learning has made it possible to model fundamental processes underlying schema encoding and storage. Yet how the brain is able to create new schemas while preserving and utilizing old schemas remains unclear. Here we propose a simple neural network framework that incorporates hierarchical gating to model the prefrontal cortex’s ability to flexibly encode and use multiple disparate schemas. We show how gating naturally leads to transfer learning and robust memory savings. We then show how neuropsychological impairments observed in patients with prefrontal damage are mimicked by lesions of our network. Our architecture, which we call DynaMoE, provides a fundamental framework for how the prefrontal cortex may handle the abundance of schemas necessary to navigate the real world.

Human memory and learning represent the most complex and miraculous of human abilities and also the most difficult and major challenges of modern medicine. While treatments for diseases from heart to lung have rapidly advanced, basic mechanisms of brain function rooted in memory and learning remain completely unknown, reflected by the dearth of targeted treatments available for neuropsychiatric disorders.This dissertation seeks to understand the neural processes of memory and learning at three interdependent levels of investigation—systems of circuits, an individual circuit, and molecular modulation of circuits—and how they can become disrupted in disease. To study these processes, I use artificial neural network models.I first consider a system of circuits (Chapter 2), studying how regions of the brain coordinate to produce coherent and flexible behavior and how disruptions lead to specific clinical syndromes. I focus on the prefrontal cortex where distinct subregions have been characterized, and analyze how simple architectural alterations in circuit interdependencies can give rise to powerful memory and learning features characteristic of humans. I then study how disruptions affect this circuit system, revealing striking parallels to patients who have suffered specific prefrontal lesions.To understand how an individual circuit learns and develops, I then focus on the internal process of strategy acquisition and evolution within a neural network (Chapter 3). To do this, I investigate a surprising experimental finding that humans and monkeys exhibit major behavioral discrepancies when solving a set of working memory tasks. I find that neural network models follow characteristic strategy progressions, exhibiting both monkey-like and human-like behavior at different stages. Furthermore, neural networks can mimic common problem solving heuristics while operating with unrelated underlying mechanisms. Counterintuitively, I find that some non-optimal heuristics can act as catalytic strategies that accelerate optimal learning.These investigations raise the question of how highly structured networks of neurons, like brains, are able to store a multitude of versatile behaviors. In Chapter 4, I investigate how a critical class of regulatory molecules in the brain called neuromodulators alter network properties to enable multiplexing of neural circuits. I reveal how neuromodulators may underlie idiosyncratic circuit dose-response properties, suggesting an alternative explanation for the large variability of neuropsychiatric drug sensitivities observed in patients.Through these three levels of investigation, I aim to contribute to our understanding of how systems, circuits, and molecules help store and manipulate forms of neural memory to generate the complex and dynamic behaviors characteristic of humans and animals. My findings suggest how these processes relate to disease manifestations and begin to suggest possible avenues toward the ultimate goal: treatments for patients.

Neuromodulators are critical controllers of neural states, with dysfunctions linked to various neuropsychiatric disorders. Although many biological aspects of neuromodulation have been studied, the computational principles underlying how neuromodulation of distributed neural populations controls brain states remain unclear. Compared with specific contextual inputs, neuromodulation is a single scalar signal that is broadcast broadly to many neurons. We model the modulation of synaptic weight in a recurrent neural network model and show that neuromodulators can dramatically alter the function of a network, even when highly simplified. We find that under structural constraints like those in brains, this provides a fundamental mechanism that can increase the computational capability and flexibility of a neural network. Diffuse synaptic weight modulation enables storage of multiple memories using a common set of synapses that are able to generate diverse, even diametrically opposed, behaviors. Our findings help explain how neuromodulators “unlock” specific behaviors by creating task-specific hyperchannels in the space of neural activities and motivate more flexible, compact and capable machine learning architectures. Neuromodulation through the release of molecules like serotonin and dopamine provides a control mechanism that allows brains to shift into distinct behavioral modes. We use an artificial neural network model to show how the action of neuromodulatory molecules acting as a broadcast signal on synaptic connections enables flexible and smooth behavioral shifting. We find that individual networks exhibit idiosyncratic sensitivities to neuromodulation under identical training conditions, highlighting a principle underlying behavioral variability. Network sensitivity is tied to the geometry of network activity dynamics, which provides an explanation for why different types of neuromodulation (molecular vs direct current modulation) have different behavioral effects. Our work suggests experiments to test biological hypotheses and paths forward in the development of flexible artificial intelligence systems.

Mood, arousal, and other internal states can drastically alter behavior, even in identical external circumstances - a cold glass of water when you are thirsty is much more desirable than when you are sated. Neuromodulators are critical controllers of such neural states, with dysfunctions linked to various neuropsychiatric disorders. Although biological aspects of neuromodulation have been well studied, the computational principles underlying how large-scale neuromodulation of distributed neural populations shifts brain states remain unclear. We use recurrent neural networks to model how synaptic weight modulation - an important function of neuromodulators - can achieve nuanced alterations in neural computation, even in a highly simplified form. We find that under structural constraints like those in brains, this provides a fundamental mechanism that can increase the computational capability and flexibility of a neural network by enabling overlapping storage of synaptic memories able to generate diverse, even diametrically opposed, behaviors. Our findings help explain how neuromodulators \"unlock\" specific behaviors by creating task-specific hypertubes in the space of neural activities and motivate more flexible, compact and capable machine learning architectures. Competing Interest Statement The authors have declared no competing interest. Footnotes * https://github.com/tsudacode/neuromodRNN

The prefrontal cortex encodes and stores numerous, often disparate, schemas and flexibly switches between them. Recent research on artificial neural networks trained by reinforcement learning has made it possible to model fundamental processes underlying schema encoding and storage. Yet how the brain is able to create new schemas while preserving and utilizing old schemas remains unclear. Here we propose a simple neural network framework based on a modification of the mixture of experts architecture to model the prefrontal cortex's ability to flexibly encode and use multiple disparate schemas. We show how incorporation of gating naturally leads to transfer learning and robust memory savings. We then show how phenotypic impairments observed in patients with prefrontal damage are mimicked by lesions of our network. Our architecture, which we call DynaMoE, provides a fundamental framework for how the prefrontal cortex may handle the abundance of schemas necessary to navigate the real world.

XSP25, previously shown to be the most abundant hydrophilic protein in xylem sap of Populus nigra in winter, belongs to a secretory protein family in which the arrangement of basic and acidic amino acids is conserved between dicotyledonous and monocotyledonous species. Its gene expression was observed at the same level in roots and shoots under long-day conditions, but highly induced under short-day conditions and at low temperatures in roots, especially in endodermis and xylem parenchyma in the root hair region of Populus trichocarpa, and its protein level was high in dormant buds, but not in roots or branches. Addition of recombinant PtXSP25 protein mitigated the denaturation of lactate dehydrogenase by drying, but showed only a slight effect on that caused by freeze–thaw cycling. Recombinant PtXSP25 protein also showed ice recrystallization inhibition activity to reduce the size of ice crystals, but had no antifreezing activity. We suggest that PtXSP25 protein produced in shoots and/or in roots under short-day conditions and at non-freezing low temperatures followed by translocation via xylem sap to shoot apoplast may protect the integrity of the plasma membrane and cell wall functions from freezing and drying damage in winter environmental conditions.