Transfer learning refers to the process of adapting a model trained on a source task to a target task. While kernel methods are conceptually and computationally simple models that are competitive on a variety of tasks, it has been unclear how to develop scalable kernel-based transfer learning methods across general source and target tasks with possibly differing label dimensions. In this work, we propose a transfer learning framework for kernel methods by projecting and translating the source model to the target task. We demonstrate the effectiveness of our framework in applications to image classification and virtual drug screening. For both applications, we identify simple scaling laws that characterize the performance of transfer-learned kernels as a function of the number of target examples. We explain this phenomenon in a simplified linear setting, where we are able to derive the exact scaling laws.

Proportional hazards models have been proposed to analyse time-to-event phenotypes in genome-wide association studies (GWAS). However, little is known about the ability of proportional hazards models to identify genetic associations under different generative models and when ascertainment is present. Here we propose the age-dependent liability threshold (ADuLT) model as an alternative to a Cox regression based GWAS, here represented by SPACox. We compare ADuLT, SPACox, and standard case-control GWAS in simulations under two generative models and with varying degrees of ascertainment as well as in the iPSYCH cohort. We find Cox regression GWAS to be underpowered when cases are strongly ascertained (cases are oversampled by a factor 5), regardless of the generative model used. ADuLT is robust to ascertainment in all simulated scenarios. Then, we analyse four psychiatric disorders in iPSYCH, ADHD, Autism, Depression, and Schizophrenia, with a strong case-ascertainment. Across these psychiatric disorders, ADuLT identifies 20 independent genome-wide significant associations, case-control GWAS finds 17, and SPACox finds 8, which is consistent with simulation results. As more genetic data are being linked to electronic health records, robust GWAS methods that can make use of age-of-onset information will help increase power in analyses for common health outcomes.

NEUROD1 is a transcription factor that helps maintain a mature phenotype of pancreatic β cells. Disruption of Neurod1 during pancreatic development causes severe neonatal diabetes; however, the exact role of NEUROD1 in the differentiation programs of endocrine cells is unknown. Here, we report a crucial role of the NEUROD1 regulatory network in endocrine lineage commitment and differentiation. Mechanistically, transcriptome and chromatin landscape analyses demonstrate that Neurod1 inactivation triggers a downregulation of endocrine differentiation transcription factors and upregulation of non-endocrine genes within the Neurod1-deficient endocrine cell population, disturbing endocrine identity acquisition. Neurod1 deficiency altered the H3K27me3 histone modification pattern in promoter regions of differentially expressed genes, which resulted in gene regulatory network changes in the differentiation pathway of endocrine cells, compromising endocrine cell potential, differentiation, and functional properties.

Chemicals or drugs can accumulate within biomolecular condensates formed through phase separation in cells. Here, we use super-resolution imaging to search for chemicals that induce phase transition within chromatin at the microscale. This microscopic screening approach reveals that adriamycin (doxorubicin) - a widely used anticancer drug that is known to interact with chromatin - specifically induces visible local condensation and global conformational change of chromatin in cancer and primary cells. Hi-C and ATAC-seq experiments systematically and quantitatively demonstrate that adriamycin-induced chromatin condensation is accompanied by weakened chromatin interaction within topologically associated domains, compartment A/B switching, lower chromatin accessibility, and corresponding transcriptomic changes. Mechanistically, adriamycin complexes with histone H1 and induces phase transition of H1, forming fibrous aggregates in vitro. These results reveal a phase separation-driven mechanism for a chemotherapeutic drug.

Epithelial-to-mesenchymal transitions (EMTs) of both endocardium and epicardium guide atrioventricular heart valve formation, but the cellular complexity and small scale of this tissue have restricted analyses. To circumvent these issues, we analyzed over 50,000 murine single-cell transcriptomes from embryonic day (E)7.75 hearts to E12.5 atrioventricular canals. We delineate mesenchymal and endocardial bifurcation during endocardial EMT, identify a distinct, transdifferentiating epicardial population during epicardial EMT, and reveal the activation of epithelial-mesenchymal plasticity during both processes. In Sox9-deficient valves, we observe increased epithelial-mesenchymal plasticity, indicating a role for SOX9 in promoting endothelial and mesenchymal cell fate decisions. Lastly, we deconvolve cell interactions guiding the initiation and progression of cardiac valve EMTs. Overall, these data reveal mechanisms of emergence of mesenchyme from endocardium or epicardium at single-cell resolution and will serve as an atlas of EMT initiation and progression with broad implications in regenerative medicine and cancer biology.

Range size is a universal characteristic of every biological species, and is often assumed to affect diversification rate. There are strong theoretical arguments that large-ranged species should have higher rates of diversification. On the other hand, the observation that small-ranged species are often phylogenetically clustered might indicate high diversification of small-ranged species. This discrepancy between theory and the data may be caused by the fact that typical methods of data analysis do not account for range size changes during speciation. Here we use a cladogenetic state-dependent diversification model applied to mammals to show that range size changes during speciation are ubiquitous and small-ranged species indeed diversify generally slower, as theoretically expected. However, both range size and diversification are strongly influenced by idiosyncratic and spatially localized events, such as colonization of an archipelago or a mountain system, which often override the general pattern of range size evolution.

There is an arms race between beta-lactam antibiotics development and co-evolving beta-lactamases, which provide resistance by breaking down beta-lactam rings. We have observed that certain beta-lactamases tend to aggregate, which persists throughout their evolution under the selective pressure of antibiotics on their active sites. Interestingly, we find that existing beta-lactamase active site inhibitors can act as molecular chaperones, promoting the proper folding of these resistance factors. Therefore, we have created Pept-Ins, synthetic peptides designed to exploit the structural weaknesses of beta-lactamases by causing them to misfold into intracellular inclusion bodies. This approach restores sensitivity to a wide range of beta-lactam antibiotics in resistant clinical isolates, including those with Extended Spectrum variants that pose significant challenges in medical practice. Our findings suggest that targeted aggregation of resistance factors could offer a strategy for identifying molecules that aid in addressing the global antibiotic resistance crisis.

Van der Waals dielectrics are fundamental materials for condensed matter physics and advanced electronic applications. Most dielectrics host isotropic structures in crystalline or amorphous forms, and only a few studies have considered the role of anisotropic crystal symmetry in dielectrics as a delicate way to tune electronic properties of channel materials. Here, we demonstrate a layered anisotropic dielectric, SiP2, with non-symmorphic twofold-rotational C2 symmetry as a gate medium which can break the original threefold-rotational C3 symmetry of MoS2 to achieve unexpected linearly-polarized photoluminescence and anisotropic second harmonic generation at SiP2/MoS2 interfaces. In contrast to the isotropic behavior of pristine MoS2, a large conductance anisotropy with an anisotropy index up to 1000 can be achieved and modulated in SiP2-gated MoS2 transistors. Theoretical calculations reveal that the anisotropic moiré potential at such interfaces is responsible for the giant anisotropic conductance and optical response. Our results provide a strategy for generating exotic functionalities at dielectric/semiconductor interfaces via symmetry engineering.

Two major earthquakes (MW 7.8 and MW 7.7) ruptured left-lateral strike-slip faults of the East Anatolian Fault Zone (EAFZ) on February 6, 2023, causing >59,000 fatalities and ~$119B in damage in southeastern Türkiye and northwestern Syria. Here we derived kinematic rupture models for the two events by inverting extensive seismic and geodetic observations using complex 5-6 segment fault models constrained by satellite observations and relocated aftershocks. The larger event nucleated on a splay fault, and then propagated bilaterally ~350 km along the main EAFZ strand. The rupture speed varied from 2.5-4.5 km/s, and peak slip was ~8.1 m. 9-h later, the second event ruptured ~160 km along the curved northern EAFZ strand, with early bilateral supershear rupture velocity (>4 km/s) followed by a slower rupture speed (~3 km/s). Coulomb Failure stress increase imparted by the first event indicates plausible triggering of the doublet aftershock, along with loading of neighboring faults.

Sarpagine alkaloids offer signicant opportunities in drug discovery, yet the efficient total syntheses and diverse structural modifications of these natural products remain highly challenging due to the architectural complexity. Here we show a homo-Mannich reaction of cyclopropanol with imines generated via a Bischler-Napieralski reaction enables a protecting-group-free, redox economic, four-step access to the tetracyclic sarpagine core from L-tryptophan esters. Based on this advancement, diversified syntheses of sarpagine alkaloids and analogues are achieved in a short synthetic route. The systematic anticancer evaluation indicates that natural products vellosimine and Na-methyl vellosimine possess modest anticancer activity. Intensive structural optimization of these lead molecules and exploration of the structure-activity relationship lead to the identification of analogue 15ai with an allene unit showing a tenfold improvement in anticancer activities. Further mechanism studies indicate compound 15ai exertes antiproliferation effects by inducing ferroptosis, which is an appealing non-apoptotic cell death form that may provide new solutions in future cancer therapies.

Genes act in concert with each other in specific contexts to perform their functions. Determining how these genes influence complex traits requires a mechanistic understanding of expression regulation across different conditions. It has been shown that this insight is critical for developing new therapies. Transcriptome-wide association studies have helped uncover the role of individual genes in disease-relevant mechanisms. However, modern models of the architecture of complex traits predict that gene-gene interactions play a crucial role in disease origin and progression. Here we introduce PhenoPLIER, a computational approach that maps gene-trait associations and pharmacological perturbation data into a common latent representation for a joint analysis. This representation is based on modules of genes with similar expression patterns across the same conditions. We observe that diseases are significantly associated with gene modules expressed in relevant cell types, and our approach is accurate in predicting known drug-disease pairs and inferring mechanisms of action. Furthermore, using a CRISPR screen to analyze lipid regulation, we find that functionally important players lack associations but are prioritized in trait-associated modules by PhenoPLIER. By incorporating groups of co-expressed genes, PhenoPLIER can contextualize genetic associations and reveal potential targets missed by single-gene strategies.

The microvasculature plays a key role in tissue perfusion and exchange of gases and metabolites. In this study we use human blood vessel organoids (BVOs) as a model of the microvasculature. BVOs fully recapitulate key features of the human microvasculature, including the reliance of mature endothelial cells on glycolytic metabolism, as concluded from metabolic flux assays and mass spectrometry-based metabolomics using stable tracing of 13C-glucose. Pharmacological targeting of PFKFB3, an activator of glycolysis, using two chemical inhibitors results in rapid BVO restructuring, vessel regression with reduced pericyte coverage. PFKFB3 mutant BVOs also display similar structural remodelling. Proteomic analysis of the BVO secretome reveal remodelling of the extracellular matrix and differential expression of paracrine mediators such as CTGF. Treatment with recombinant CTGF recovers microvessel structure. In this work we demonstrate that BVOs rapidly undergo restructuring in response to metabolic changes and identify CTGF as a critical paracrine regulator of microvascular integrity.

Silicon-stereogenic optically active silylboranes could potentially allow the formation of chiral silyl nucleophiles as well as the synthesis of various chiral silicon compounds. However, the synthesis of such silicon-stereogenic silylboranes has not been achieved so far. Here, we report the synthesis of silicon-stereogenic optically active silylboranes via a stereospecific Pt(PPh3)4-catalyzed Si-H borylation of chiral hydrosilanes, which are synthesized by stoichiometric and catalytic asymmetric synthesis, in high yield and very high or perfect enantiospecificity (99% es in one case, and >99% es in the others) with retention of the configuration. Furthermore, we report a practical approach to generate silicon-stereogenic silyl nucleophiles with high enantiopurity and configurational stability using MeLi activation. This protocol is suitable for the stereospecific and general synthesis of silicon-stereogenic trialkyl-, dialkylbenzyl-, dialkylaryl-, diarylalkyl-, and alkylary benzyloxy-substituted silylboranes and their corresponding silyl nucleophiles with excellent enantiospecificity (>99% es except one case of 99% es). Transition-metal-catalyzed C-Si bond-forming cross-coupling reactions and conjugate-addition reactions are also demonstrated. The mechanisms underlying the stability and reactivity of such chiral silyl anion were investigated by combining NMR spectroscopy and DFT calculations.

Certain times of major biotic replacement have often been interpreted as broadly competitive, mediated by innovation in the succeeding clades. A classic example was the switch from brachiopods to bivalves as major seabed organisms following the Permian-Triassic mass extinction (PTME), ~252 million years ago. This was attributed to competitive exclusion of brachiopods by the better adapted bivalves or simply to the fact that brachiopods had been hit especially hard by the PTME. The brachiopod-bivalve switch is emblematic of the global turnover of marine faunas from Palaeozoic-type to Modern-type triggered by the PTME. Here, using Bayesian analyses, we find that unexpectedly the two clades displayed similar large-scale trends of diversification before the Jurassic. Insight from a multivariate birth-death model shows that the extinction of major brachiopod clades during the PTME set the stage for the brachiopod-bivalve switch, with differential responses to high ocean temperatures post-extinction further facilitating their displacement by bivalves. Our study strengthens evidence that brachiopods and bivalves were not competitors over macroevolutionary time scales, with extinction events and environmental stresses shaping their divergent fates.

An important challenge in genetics, evolution and biotechnology is to understand and predict how mutations combine to alter phenotypes, including molecular activities, fitness and disease. In diploids, mutations in a gene can combine on the same chromosome or on different chromosomes as a "heteroallelic combination". However, a direct comparison of the extent, sign, and stability of the genetic interactions between variants within and between alleles is lacking. Here we use thermodynamic models of protein folding and ligand-binding to show that interactions between mutations within and between alleles are expected in even very simple biophysical systems. Protein folding alone generates within-allele interactions and a single molecular interaction is sufficient to cause between-allele interactions and dominance. These interactions change differently, quantitatively and qualitatively as a system becomes more complex. Altering the concentration of a ligand can, for example, switch alleles from dominant to recessive. Our results show that intra-molecular epistasis and dominance should be widely expected in even the simplest biological systems but also reinforce the view that they are plastic system properties and so a formidable challenge to predict. Accurate prediction of both intra-molecular epistasis and dominance will require either detailed mechanistic understanding and experimental parameterization or brute-force measurement and learning.

Graphene has aroused great attention due to the intriguing properties associated with its low-energy Dirac Hamiltonian. When graphene is coupled with a correlated insulating substrate, electronic states that cannot be revealed in either individual layer may emerge in a synergistic manner. Here, we theoretically study the correlated and topological states in Coulomb-coupled and gate-tunable graphene-insulator heterostructures. By electrostatically aligning the electronic bands, charge carriers transferred between graphene and the insulator can yield a long-wavelength electronic crystal at the interface, exerting a superlattice Coulomb potential on graphene and generating topologically nontrivial subbands. This coupling can further boost electron-electron interaction effects in graphene, leading to a spontaneous bandgap formation at the Dirac point and interaction-enhanced Fermi velocity. Reciprocally, the electronic crystal at the interface is substantially stabilized with the help of cooperative interlayer Coulomb coupling. We propose a number of substrate candidates for graphene to experimentally demonstrate these effects.

Many strongly correlated transition metal insulators are colored, even though they have band gaps much larger than the highest energy photons from the visible light. An adequate explanation for the color requires a theoretical approach able to compute subgap excitons in periodic crystals, reliably and without free parameters-a formidable challenge. The literature often fails to disentangle two important factors: what makes excitons form and what makes them optically bright. We pick two archetypal cases as examples: NiO with green color and MnF2 with pink color, and employ two kinds of ab initio many body Green's function theories; the first, a perturbative theory based on low-order extensions of the GW approximation, is able to explain the color in NiO, while the same theory is unable to explain why MnF2 is pink. We show its color originates from higher order spin-flip transitions that modify the optical response, which is contained in dynamical mean-field theory (DMFT). We show that symmetry lowering mechanisms may determine how 'bright' these excitons are, but they are not fundamental to their existence.

The introduction of metal sites into molecular metal oxides, so-called polyoxometalates, is key for tuning their structure and reactivity. The complex mechanisms which govern metal-functionalization of polyoxometalates are still poorly understood. Here, we report a coupled set of light-dependent and light-independent reaction equilibria controlling the mono- and di-metal-functionalization of a prototype molecular vanadium oxide cluster. Comprehensive mechanistic analyses show that coordination of a Mg2+ ion to the species {(NMe2H2)2[VV12O32Cl]}3- results in formation of the mono-functionalized {(NMe2H2)[(MgCl)VV12O32Cl]}3- with simultaneous release of a NMe2H2+ placeholder cation. Irradiation of this species with visible light results in one-electron reduction of the vanadate, exchange of the second NMe2H2+ with Mg2+, and formation/crystallization of the di-metal-functionalized [(MgCl)2VIVVV11O32Cl]4-. Mechanistic studies show how stimuli such as light or competing cations affect the coupled equilibria. Transfer of this synthetic concept to other metal cations is also demonstrated, highlighting the versatility of the approach.

Undulatory swimming is the predominant form of locomotion in aquatic vertebrates. A myriad of animals of different species and sizes oscillate their bodies to propel themselves in aquatic environments with swimming speed scaling as the product of the animal length by the oscillation frequency. Although frequency tuning is the primary means by which a swimmer selects its speed, there is no consensus on the mechanisms involved. In this article, we propose scaling laws for undulatory swimmers that relate oscillation frequency to length by taking into account both the biological characteristics of the muscles and the interaction of the moving swimmer with its environment. Results are supported by an extensive literature review including approximately 1200 individuals of different species, sizes and swimming environments. We highlight a crossover in size around 0.5-1 m. Below this value, the frequency can be tuned between 2-20 Hz due to biological constraints and the interplay between slow and fast muscles. Above this value, the fluid-swimmer interaction must be taken into account and the frequency is inversely proportional to the length of the animal. This approach predicts a maximum swimming speed around 5-10 m.s-1 for large swimmers, consistent with the threshold to prevent bubble cavitation.

Carbon supports are ubiquitous components of heterogeneous catalysts for acetylene hydrochlorination to vinyl chloride, from commercial mercury-based systems to more sustainable metal single-atom alternatives. Their potential co-catalytic role has long been postulated but never unequivocally demonstrated. Herein, we evidence the bifunctionality of carbons and metal sites in the acetylene hydrochlorination catalytic cycle. Combining operando X-ray absorption spectroscopy with other spectroscopic and kinetic analyses, we monitor the structure of single metal atoms (Pt, Au, Ru) and carbon supports (activated, non-activated, and nitrogen-doped) from catalyst synthesis, using various procedures, to operation at different conditions. Metal atoms exclusively activate hydrogen chloride, while metal-neighboring sites in the support bind acetylene. Resolving the coordination environment of working metal atoms guides theoretical simulations in proposing potential binding sites for acetylene in the support and a viable reaction profile. Expanding from single-atom to ensemble catalysis, these results reinforce the importance of optimizing both metal and support components to leverage the distinct functions of each for advancing catalyst design.