Search

The search for macroscopic quantum phenomena is a fundamental pursuit inquantum mechanics. It allows us to test the limits quantum physics and provides new avenues for exploring the interplay between quantum mechanics and relativity. In this work, we introduce a novel approach to generate macroscopic quantum systems by demonstrating that the creation process of a quantum system can span a macroscopic distance. Specifically, we generate photon pairs in a coherent superposition of two origins separated by up to 70 meters. This newapproach not only provides an exciting opportunity for foundational experimentsin quantum physics, but also has practical applications for high-precision measurements of distributed properties such as pressure and humidity of air or gases.

String-based molecular representations play a crucial role in cheminformatics applications, and with the growing success of deep learning in chemistry, have been readily adopted into machine learning pipelines. However, traditionalstring-based representations such as SMILES are often prone to syntactic and semantic errors when produced by generative models. To address these problems, a novel representation, SELF-referencIng Embedded Strings (SELFIES), was proposed that is inherently 100% robust, alongside an accompanying open-source implementation. Since then, we have generalized SELFIES to support a wider range of molecules and semantic constraints and streamlined its underlying grammar. We have implemented this updated representation in subsequent versionsof \selfieslib, where we have also made major advances with respect to design, efficiency, and supported features. Hence, we present the current status of \selfieslib (version 2.1.1) in this manuscript.

Photons are the physical system of choice for performing experimental tests of the foundations of quantum mechanics. Furthermore, photonic quantum technology is a main player in the second quantum revolution, promising the development of better sensors, secure communications, and quantum-enhanced computation. These endeavors require generating specific quantum states or efficiently performing quantum tasks. The design of the corresponding optical experiments was historically powered by human creativity but is recently being automated with advanced computer algorithms and artificial intelligence. While several computer-designed experiments have been experimentally realized, this approach has not yet been widely adopted by the broader photonic quantum optics community. The main roadblocks consist of most systems being closed-source, inefficient, or targeted to very specific use-cases that are difficult to generalize. Here, we overcome these problems with a highly-efficient, open-source digital discovery framework PyTheus, which can employ a wide range of experimental devices from modern quantum labs to solve various tasks. This includes the discovery of highly entangled quantum states, quantum measurement schemes, quantum communication protocols, multi-particle quantum gates, as well as the optimization of continuous and discrete properties of quantum experiments or quantum states. PyTheus produces interpretable designs for complex experimental problems which human researchers can often readily conceptualize. PyTheus is an example of a powerful framework that can lead to scientific discoveries -- one of the core goals of artificial intelligence in science. We hope it will help accelerate the development of quantum optics and provide new ideas in quantum hardware and technology.

Entanglement is a crucial resource for quantum technologies ranging from quantum communication to quantum-enhanced measurements and computation. Finding experimental setups for these tasks is a conceptual challenge for human scientists due to the counterintuitive behavior of multiparticle interference and the enormously large combinatorial search space. Recently, new possibilities have been opened by artificial discovery where artificial intelligence proposes experimental setups for the creation and manipulation of high-dimensional multi-particle entanglement. While digitally discovered experiments go beyond what has been conceived by human experts, a crucial goal is to understand the underlying concepts which enable these new useful experimental blueprints. Here, we present Halo (Hyperedge Assembly by Linear Optics), a new form of multiphoton quantum interference with surprising properties. Halos were used by our digital discovery framework to solve previously open questions. We -- the human part of this collaboration -- were then able to conceptualize the idea behind the computer discovery and describe them in terms of effective probabilistic multi-photon emitters. We then demonstrate its usefulness as a core of new experiments for highly entangled states, communication in quantum networks, and photonic quantum gates. Our manuscript has two conclusions. First, we introduce and explain the physics of a new practically useful multi-photon interference phenomenon that can readily be realized in advanced setups such as integrated photonic circuits. Second, our manuscript demonstrates how artificial intelligence can act as a source of inspiration for the scientific discoveries of new actionable concepts in physics.

A tool that could suggest new personalized research directions and ideas by taking insights from the scientific literature could significantly accelerate the progress of science. A field that might benefit from such an approach is artificial intelligence (AI) research, where the number of scientific publications has been growing exponentially over the last years, making it challenging for human researchers to keep track of the progress. Here, we use AI techniques to predict the future research directions of AI itself. We develop a new graph-based benchmark based on real-world data -- the Science4Cast benchmark, which aims to predict the future state of an evolving semantic network of AI. For that, we use more than 100,000 research papers and build up a knowledge network with more than 64,000 concept nodes. We then present ten diverse methods to tackle this task, ranging from pure statistical to pure learning methods. Surprisingly, the most powerful methods use acarefully curated set of network features, rather than an end-to-end AI approach. It indicates a great potential that can be unleashed for purely ML approaches without human knowledge. Ultimately, better predictions of new future research directions will be a crucial component of more advanced research suggestion tools.

ObjectiveWe aimed to identify genes associated with genetic generalized epilepsy (GGE) by combining large cohorts enriched with individuals with a positive family history. Secondarily, we set out to compare the association of genes independently with familial and sporadic GGE.MethodsWe performed a case-control whole exome sequencing study in unrelated individuals of European descent diagnosed with GGE (previously recruited and sequenced through multiple international collaborations) and ancestry-matched controls. The association of ultra-rare variants (URVs; in 18834 protein-coding genes) with epilepsy was examined in 1928 individuals with GGE (vs. 8578 controls), then separately in 945 individuals with familial GGE (vs. 8626 controls), and finally in 1005 individuals with sporadic GGE (vs. 8621 controls). We additionally examined the association of URVs with familial and sporadic GGE in two gene sets important for inhibitory signaling (19genes encoding γ-aminobutyric acid type A [GABAA ] receptors, 113genes representing the GABAergic pathway).ResultsGABRG2 was associated with GGE (p=1.8×10-5 ), approaching study-wide significance in familial GGE (p=3.0×10-6 ), whereas no gene approached a significant association with sporadic GGE. Deleterious URVs in the most intolerant subgenic regions in genes encoding GABAA receptors were associated with familial GGE (odds ratio [OR]=3.9, 95% confidence interval [CI]=1.9-7.8, false discovery rate [FDR]-adjusted p=.0024), whereas their association with sporadic GGE had marginally lower odds (OR=3.1, 95% CI=1.3-6.7, FDR-adjusted p=.022). URVs in GABAergic pathway genes were associated with familial GGE (OR=1.8, 95% CI=1.3-2.5, FDR-adjusted p=.0024) but not with sporadic GGE (OR=1.3, 95% CI=.9-1.9, FDR-adjusted p=.19).SignificanceURVs in GABRG2 are likely an important risk factor for familial GGE. The association of gene sets of GABAergic signaling with familial GGE is more prominent than with sporadic GGE.

Logic artificial intelligence (AI) is a subfield of AI where variables can take two defined arguments, True or False, and are arranged in clauses that follow the rules of formal logic. Several problems that span from physical systems to mathematical conjectures can be encoded into these clauses and be solved by checking their satisfiability (SAT). Recently, SAT solvers have become a sophisticated and powerful computational tool capable, among other things, of solving long-standing mathematical conjectures. In this work, we propose the use of logic AI for the design of optical quantum experiments. We show how to map into a SAT problem the experimental preparation of an arbitrary quantum state and propose a logic-based algorithm, called Klaus, to find an interpretable representation of the photonic setup that generates it. We compare the performance of Klaus with the state-of-the-art algorithm for this purpose based on continuous optimization. We also combine both logic and numeric strategies to find that the use of logic AI improves significantly the resolution of this problem, paving the path to develop more formal-based approaches in the context of quantum physics experiments.

Annals of Clinical and Translational Neurology 8(7), 1376-1387 (2021). doi:10.1002/acn3.51374, Published by Wiley, Chichester [u.a.]

OBJECTIVE: The mechanisms by which antiepileptic drugs (AEDs) cause birth defects (BDs) are unknown. Data suggest that AED-induced BDs may result from a genome-wide increase of de novo variants in the embryo, a mechanism which we investigated. METHODS: Whole-exome sequencing data from child-parent trios were interrogated for de novo single-nucleotide variants/indels (dnSNVs/indels) and copy number variants (dnCNVs). Generalized linear models were applied to assess de novo variant burdens in: children exposed prenatally to AEDs (AED-exposed children) vs children without BDs not exposed prenatally to AEDs (AED-unexposed unaffected children), and AED-exposed children with BDs vs those without BDs, adjusting for confounders. Fisher's exact test was used to compare categorical data. RESULTS: 67 child-parent trios were included: 10 with AED-exposed children with BDs; 46 with AED-exposed unaffected children; 11 with AED-unexposed unaffected children. The dnSNV/indel burden did not differ between AED-exposed children and AED-unexposed unaffected children [median dnSNV/indel number/child (range): 3 (0-7) vs 3 (1-5), p = 0.50]. Among AED-exposed children, there were no significant differences between those with BDs and those unaffected. Likely deleterious dnSNVs/indels were detected in 9/67 (13%) children, none of whom had BDs. The proportion of cases harbouring likely deleterious dnSNVs/indels did not differ significantly between AED-unexposed and AED-exposed children. The dnCNV burden was not associated with AED exposure or birth outcome. INTERPRETATION: Our study indicates that prenatal AED exposure does not increase the burden of de novo variants, and that this mechanism is not a major contributor to AED-induced BDs. These results can be incorporated in routine patient counselling. This article is protected by copyright. All rights reserved.

Masking is the best-researched countermeasure against side-channel analysis attacks. Even though masking was introduced almost 20 years ago, its efficient implementation continues to be an active research topic. Many of the existing works focus on the reduction of randomness requirements since the production of fresh random bits with high entropy is very costly in practice. Most of these works rely on the assumption that only so-called online randomness results in additional costs. In practice, however, it shows that the distinction between randomness costs to produce the initial masking and the randomness to maintain security during computation (online) is not meaningful. In this work, we thus study the question of minimum randomness requirements for first-order Boolean masking when taking the costs for initial randomness into account. We demonstrate that first-order masking can in theory always be performed by just using two fresh random bits and without requiring online randomness. We first show that two random bits are enough to mask linear transformations and then discuss prerequisites under which nonlinear transformations are securely performed likewise. Subsequently, we introduce a new masked AND gate that fulfills these requirements and which forms the basis for our synthesis tool that automatically transforms an unmasked implementation into a first-order secure masked implementation. We demonstrate the feasibility of this approach by implementing AES in software with only two bits of randomness, including the initial masking. Finally, we use these results to discuss the gap between theory and practice and the need for more accurate adversary models.

Objective: To ascertain the clinical and genetic factors contributing to carbamazepine- and oxcarbazepine-induced hyponatremia (COIH), and to carbamazepine (CBZ) metabolism, in a retrospectively collected, cross-sectional cohort of people with epilepsy. Methods: We collected data on serum sodium levels and antiepileptic drug levels in people with epilepsy attending a tertiary epilepsy center while on treatment with CBZ or OXC. We defined hyponatremia as Na+ ≤134 mEq/L. We estimated the CBZ metabolic ratio defined as the log transformation of the ratio of metabolite CBZ-diol to unchanged drug precursor substrate as measured in serum. Results: Clinical and genetic data relating to carbamazepine and oxcarbazepine trials were collected in 1141 patients. We did not observe any genome-wide significant associations with sodium level in a linear trend or hyponatremia as a dichotomous trait. Age, sex, number of comedications, phenytoin use, phenobarbital use, and sodium valproate use were significant predictors of CBZ metabolic ratio. No genome-wide significant associations with CBZ metabolic ratio were found. Significance: Although we did not detect a genetic predictor of hyponatremia or CBZ metabolism in our cohort, our findings suggest that the determinants of CBZ metabolism are multifactorial., SCOPUS: ar.j, info:eu-repo/semantics/published

Challenges in natural sciences can often be phrased as optimization problems. Machine learning techniques have recently been applied to solve such problems. One example in chemistry is the design of tailor-made organic materials and molecules, which requires efficient methods to explore the chemical space. We present a genetic algorithm (GA) that is enhanced with a neural network (DNN) based discriminator model to improve the diversity of generated molecules and at the same time steer the GA. We show that our algorithm outperforms other generative models in optimization tasks. We furthermore present a way to increase interpretability of genetic algorithms, which helped us to derive design principles., Comment: 9+3 Pages, 7+4 figures, 2 tables. Comments are welcome! (code is available at: https://github.com/aspuru-guzik-group/GA)

Ataxia-telangiectasia (A-T) is an autosomal recessive disorder caused by mutations in ATM, encoding a serine-threonine protein kinase that is crucially involved in DNA repair mechanisms. Clinical features include cerebellar degeneration, telangiectasia, immunodeficiency, and an increased risk of malignancies.(1) The classic form of A-T is characterized by infantile, rapidly progressing neurodegeneration and can be differentiated from variant A-T with a comparably milder disease course.(2,3) However, only a tiny fraction of patients first present with symptoms in adulthood.(4) The broad phenotypic spectrum of A-T now becomes gradually disentangled owing to the increased availability of comprehensive genetic testing.(5)

SummaryObjective Mesial temporal lobe epilepsy with hippocampal sclerosis (MTLE-HS) is a common epilepsy syndrome that is often poorly controlled by antiepileptic drug (AED) treatment. Comparative AED effectiveness studies in this condition are lacking. We report retention, efficacy, and tolerability in a cohort of patients with MTLE-HS. Methods Clinical data were collected from a European database of patients with epilepsy. We estimated retention, 12-month seizure freedom, and adverse drug reaction (ADR) rates for the 10 most commonly used AEDs in patients with MTLE-HS. Results Seven hundred sixty-seven patients with a total of 3,249 AED trials were included. The highest 12-month retention rates were observed with carbamazepine (85.9%), valproate (85%), and clobazam (79%). Twelve-month seizure freedom rates varied from 1.2% for gabapentin and vigabatrin to 11% for carbamazepine. Response rates were highest for AEDs that were prescribed as initial treatment and lowest for AEDs that were used in a third or higher instance. ADRs were reported in 47.6% of patients, with the highest rates observed with oxcarbazepine (35.7%), topiramate (30.9%), and pregabalin (27.4%), and the lowest rates with clobazam (6.5%), gabapentin (8.9%), and lamotrigine (16.6%). The most commonly reported ADRs were lethargy and drowsiness, dizziness, vertigo and ataxia, and blurred vision and diplopia. Significance Our results did not demonstrate any clear advantage of newer versus older AEDs. Our results provide useful insights into AED retention, efficacy, and ADR rates in patients with MTLE-HS.