RESUMO
Artificial intelligence (AI) is defined as the ability of machines to perform tasks that are usually associated with intelligent beings. Argument and debate are fundamental capabilities of human intelligence, essential for a wide range of human activities, and common to all human societies. The development of computational argumentation technologies is therefore an important emerging discipline in AI research1. Here we present Project Debater, an autonomous debating system that can engage in a competitive debate with humans. We provide a complete description of the system's architecture, a thorough and systematic evaluation of its operation across a wide range of debate topics, and a detailed account of the system's performance in its public debut against three expert human debaters. We also highlight the fundamental differences between debating with humans as opposed to challenging humans in game competitions, the latter being the focus of classical 'grand challenges' pursued by the AI research community over the past few decades. We suggest that such challenges lie in the 'comfort zone' of AI, whereas debating with humans lies in a different territory, in which humans still prevail, and for which novel paradigms are required to make substantial progress.
Assuntos
Inteligência Artificial , Comportamento Competitivo , Dissidências e Disputas , Atividades Humanas , Inteligência Artificial/normas , Humanos , Processamento de Linguagem NaturalRESUMO
Importance: With the continuous increase in the prevalence of autistic spectrum disorder (ASD), effective early screening is crucial for initiating timely interventions and improving outcomes. Objective: To develop predictive models for ASD using routinely collected developmental surveillance data and to assess their performance in predicting ASD at different ages and in different clinical scenarios. Design, Setting, and Participants: This retrospective cohort study used nationwide data of developmental assessments conducted between January 1, 2014, and January 17, 2023, with minimal follow-up of 4 years and outcome collection in March 2023. Data were from a national program of approximately 1000 maternal child health clinics that perform routine developmental surveillance of children from birth to 6 years of age, serving 70% of children in Israel. The study included all children who were assessed at the maternal child health clinics (N = 1â¯187â¯397). Children were excluded if they were born at a gestational age of 33 weeks or earlier, had no record of gestational age, or were followed up for less than 4 years without an ASD outcome. The data set was partitioned at random into a development set (80% of the children) and a holdout evaluation set (20% of the children), both with the same prevalence of ASD outcome. Exposures: For each child, demographic and birth-related covariates were extracted, as were per-visit growth measurements, quantified developmental milestone assessments, and referral summary covariates. Only information that was available before the prediction age was used for training and evaluating the models. Main Outcome and Measure: The main outcome was eligibility for a governmental disabled child allowance due to ASD, according to administrative data of the National Insurance Institute of Israel. The performance of the models that predict the outcome was evaluated and compared with previous work on the Modified Checklist for Autism in Toddlers (M-CHAT). Results: The study included 1â¯187â¯397 children (610â¯588 [51.4%] male). The performance of the ASD prediction models improved with prediction age, with fair accuracy already at 12 months of age. A model that combined longitudinal measures of developmental milestone assessments with a minimal set of demographic variables, which was applied at 18 to 24 months of age, achieved an area under the receiver operating characteristic curve of 0.83, with a sensitivity of 45.1% at a specificity of 95.0%. A model using single-visit assessments achieved an area under the receiver operating characteristic curve of 0.81 and a sensitivity of 41.2% at a specificity of 95.0%. The best performing prediction models surpassed the pooled performance of M-CHAT (sensitivity, 40%; specificity, 95%) reported in studies with a similar design. Conclusions and Relevance: This cohort study found that ASD can be predicted from routine developmental surveillance data at an accuracy surpassing M-CHAT screening. This tool may be seamlessly integrated in the clinical workflow to improve early identification of children who may benefit from timely interventions.
Assuntos
Transtorno do Espectro Autista , Transtorno Autístico , Criança , Pré-Escolar , Feminino , Humanos , Lactente , Masculino , Transtorno do Espectro Autista/diagnóstico , Transtorno do Espectro Autista/epidemiologia , Transtorno Autístico/diagnóstico , Transtorno Autístico/epidemiologia , Estudos de Coortes , Estudos RetrospectivosRESUMO
PURPOSE: Youth mental distress has substantially increased during the COVID-19 pandemic. However, it is unclear if mental symptoms are directly related to SARS-CoV-2 infection or to social restrictions. We aimed to investigate mental health outcomes in infected versus uninfected adolescents, for up to two years after an index polymerase chain reaction (PCR) test. METHODS: A retrospective cohort study, based on electronic health records from a large nationally representative Israeli health fund, among adolescents aged 12-17 years with a PCR test for SARS-CoV-2 between March 1, 2020 and March 1, 2021. Infected and uninfected individuals were matched by age, sex, test date, sector, and socioeconomic status. Cox regression was used to derive hazard ratios (HRs) for mental health outcomes within two years from PCR test for infected versus uninfected individuals, while accounting for pre-existing psychiatric history. External validation was performed on UK primary care data. RESULTS: Among 146,067 PCR-tested adolescents, 24,009 were positive and 22,354 were matched with negative adolescents. SARS-CoV-2 infection was significantly associated with reduced risks for dispensation of antidepressants (HR 0.74, 95% confidence interval [CI] 0.66-0.83), diagnoses of anxiety (HR 0.82, 95% CI 0.71-0.95), depression (HR 0.65, 95% CI 0.53-0.80), and stress (HR 0.80, 95% CI 0.69-0.92). Similar results were obtained in the validation dataset. DISCUSSION: This large, population-based study suggests that SARS-CoV-2 infection is not associated with elevated risk for mental distress in adolescents. Our findings highlight the importance of taking a holistic view on adolescents' mental health during the pandemic, with consideration of both SARS-CoV-2 infection and response measures.
Assuntos
COVID-19 , Síndrome de COVID-19 Pós-Aguda , Adolescente , Humanos , COVID-19/epidemiologia , Pandemias , Estudos Retrospectivos , SARS-CoV-2 , Avaliação de Resultados em Cuidados de SaúdeRESUMO
BACKGROUND: Depression is a major cause of disability worldwide. Recent data suggest that, in industrialised countries, the prevalence of depression peaks in middle age. Identifying factors predictive of future depressive episodes is crucial for developing prevention strategies for this age group. AIMS: We aimed to identify future depression in middle-aged adults with no previous psychiatric history. METHOD: To predict a diagnosis of depression 1 year or more following a comprehensive baseline assessment, we used a data-driven, machine-learning methodology. Our data-set was the UK Biobank of middle-aged participants (N = 245 036) with no psychiatric history. RESULTS: Overall, 2.18% of the study population developed a depressive episode at least 1 year following baseline. Basing predictions on a single mental health questionnaire led to an area under the curve of the receiver operating characteristic of 0.66, and a predictive model leveraging the combined results of 100 UK Biobank questionnaires and measurements improved this to 0.79. Our findings were robust to demographic variations (place of birth, gender) and variations in methods of depression assessment. Thus, machine-learning-based models best predict diagnoses of depression when allowing the inclusion of multiple features. CONCLUSIONS: Machine-learning approaches show potential for being beneficial for the identification of clinically relevant predictors of depression. Specifically, we can identify, with moderate success, people with no recorded psychiatric history as at risk for depression by using a relatively small number of features. More work is required to improve these models and evaluate their cost-effectiveness before integrating them into the clinical workflow.
RESUMO
BACKGROUND: Developmental surveillance, conducted routinely worldwide, is fundamental for timely identification of children at risk of developmental delays. It is typically executed by assessing age-appropriate milestone attainment and applying clinical judgment during health supervision visits. Unlike developmental screening and evaluation tools, surveillance typically lacks standardized quantitative measures, and consequently, its interpretation is often qualitative and subjective. OBJECTIVE: Herein, we suggested a novel method for aggregating developmental surveillance assessments into a single score that coherently depicts and monitors child development. We described the procedure for calculating the score and demonstrated its ability to effectively capture known population-level associations. Additionally, we showed that the score can be used to describe longitudinal patterns of development that may facilitate tracking and classifying developmental trajectories of children. METHODS: We described the Developmental Surveillance Score (DSS), a simple-to-use tool that quantifies the age-dependent severity level of a failure at attaining developmental milestones based on the recently introduced Israeli developmental surveillance program. We evaluated the DSS using a nationwide cohort of >1 million Israeli children from birth to 36 months of age, assessed between July 1, 2014, and September 1, 2021. We measured the score's ability to capture known associations between developmental delays and characteristics of the mother and child. Additionally, we computed series of the DSS in consecutive visits to describe a child's longitudinal development and applied cluster analysis to identify distinct patterns of these developmental trajectories. RESULTS: The analyzed cohort included 1,130,005 children. The evaluation of the DSS on subpopulations of the cohort, stratified by known risk factors of developmental delays, revealed expected relations between developmental delay and characteristics of the child and mother, including demographics and obstetrics-related variables. On average, the score was worse for preterm children compared to full-term children and for male children compared to female children, and it was correspondingly worse for lower levels of maternal education. The trajectories of scores in 6 consecutive visits were available for 294,000 children. The clustering of these trajectories revealed 3 main types of developmental patterns that are consistent with clinical experience: children who successfully attain milestones, children who initially tend to fail but improve over time, and children whose failures tend to increase over time. CONCLUSIONS: The suggested score is straightforward to compute in its basic form and can be easily implemented as a web-based tool in its more elaborate form. It highlights known and novel relations between developmental delay and characteristics of the mother and child, demonstrating its potential usefulness for surveillance and research. Additionally, it can monitor the developmental trajectory of a child and characterize it. Future work is needed to calibrate the score vis-a-vis other screening tools, validate it worldwide, and integrate it into the clinical workflow of developmental surveillance.
Assuntos
Desenvolvimento Infantil , Pré-Escolar , Feminino , Humanos , Masculino , Gravidez , Valores de ReferênciaRESUMO
OBJECTIVE: Adolescents' mental health was severely compromised during the COVID-19 pandemic. Longitudinal real-world studies on changes in the mental health of adolescents during the later phase of the pandemic are limited. We aimed to quantify the effect of COVID-19 pandemic on adolescents' mental health outcomes based on electronic health records. METHOD: This was a retrospective cohort study using the computerized database of a 2.5 million members, state-mandated health organization in Israel. Rates of mental health diagnoses and psychiatric drug dispensations were measured among adolescents 12 to 17 years of age with and without pre-existing mental history, for the years 2017 to 2021. Relative risks were computed between the years, and interrupted time series (ITS) analyses evaluated changes in monthly incidence rates of psychiatric outcomes. RESULTS: The average population size was 218,146 in 2021. During the COVID-19 period, a 36% increase was observed in the incidence of depression (95% CI = 25-47), 31% in anxiety (95% CI = 23-39), 20% in stress (95% CI = 13-27), 50% in eating disorders (95% CI = 35-67), 25% in antidepressant use (95% CI = 25-33), and 28% in antipsychotic use (95% CI = 18-40). A decreased rate of 26% (95% CI = 0.80-0.88) was observed in ADHD diagnoses. The increase of the examined outcomes was most prominent among youth without psychiatric history, female youth, general secular Jewish population, youth with medium-high socioeconomic status, and those 14 to 15 years of age. ITS analysis confirmed a significantly higher growth in the incidence of psychiatric outcomes during the COVID-19 period, compared to those in previous years. CONCLUSION: This real-world study highlights the deterioration of adolescents' mental health during the COVID-19 pandemic and suggests that youth mental health should be considered during health policy decision making. DIVERSITY & INCLUSION STATEMENT: We worked to ensure sex and gender balance in the recruitment of human participants. We worked to ensure race, ethnic, and/or other types of diversity in the recruitment of human participants. We actively worked to promote sex and gender balance in our author group. The author list of this paper includes contributors from the location and/or community where the research was conducted who participated in the data collection, design, analysis, and/or interpretation of the work.
Assuntos
Antipsicóticos , COVID-19 , Masculino , Humanos , Adolescente , Feminino , Saúde Mental , COVID-19/epidemiologia , Pandemias , Estudos RetrospectivosRESUMO
Comparative analysis is a fundamental tool in biology. Conservation among species greatly assists the detection and characterization of functional elements, whereas inter-species differences are probably the best indicators of biological adaptation. Traditionally, comparative approaches were applied to the analysis of genomic sequences. With the growing availability of functional genomic data, comparative paradigms are now being extended also to the study of other functional attributes, most notably the gene expression. Here we review recent works applying comparative analysis to large-scale gene expression datasets and discuss the central principles and challenges of such approaches.
Assuntos
Perfilação da Expressão Gênica/métodos , Genômica/métodos , Animais , Biologia Computacional/métodos , Humanos , Modelos Teóricos , Análise de Sequência de DNA/métodos , Especificidade da EspécieRESUMO
Variation in gene expression levels on a genomic scale has been detected among different strains, among closely related species, and within populations of genetically identical cells. What are the driving forces that lead to expression divergence in some genes and conserved expression in others? Here we employ flux balance analysis to address this question for metabolic genes. We consider the genome-scale metabolic model of Saccharomyces cerevisiae, and its entire space of optimal and near-optimal flux distributions. We show that this space reveals underlying evolutionary constraints on expression regulation, as well as on the conservation of the underlying gene sequences. Genes that have a high range of optimal flux levels tend to display divergent expression levels among different yeast strains and species. This suggests that gene regulation has diverged in those parts of the metabolic network that are less constrained. In addition, we show that genes that are active in a large fraction of the space of optimal solutions tend to have conserved sequences. This supports the possibility that there is less selective pressure to maintain genes that are relevant for only a small number of metabolic states.
Assuntos
Regulação Fúngica da Expressão Gênica/fisiologia , Modelos Biológicos , Mapeamento de Interação de Proteínas/métodos , Proteínas de Saccharomyces cerevisiae/fisiologia , Saccharomyces cerevisiae/fisiologia , Transdução de Sinais/fisiologia , Proliferação de Células , Simulação por ComputadorRESUMO
Multiple Sequence Alignment (MSA) is one of the most fundamental problems in computational molecular biology. The running time of the best known scheme for finding an optimal alignment, based on dynamic programming, increases exponentially with the number of input sequences. Hence, many heuristics were suggested for the problem. We consider a version of the MSA problem where the goal is to find an optimal alignment in which matches are restricted to positions in predefined matching segments. We present several techniques for making the dynamic programming algorithm more efficient, while still finding an optimal solution under these restrictions. We prove that it suffices to find an optimal alignment of the predefined sequence segments, rather than single letters, thereby reducing the input size and thus improving the running time. We also identify "shortcuts" that expedite the dynamic programming scheme. Empirical study shows that, taken together, these observations lead to an improved running time over the basic dynamic programming algorithm by 4 to 12 orders of magnitude, while still obtaining an optimal solution. Under the additional assumption that matches between segments are transitive, we further improve the running time for finding the optimal solution by restricting the search space of the dynamic programming algorithm.
Assuntos
Algoritmos , Mapeamento Cromossômico/métodos , Elementos de DNA Transponíveis/genética , Análise de Sequência de DNA/métodos , Análise de Sequência de Proteína/métodos , Pareamento Incorreto de Bases , Alinhamento de Sequência/métodosRESUMO
The ProtoNet site provides an automatic hierarchical clustering of the SWISS-PROT protein database. The clustering is based on an all-against-all BLAST similarity search. The similarities' E-score is used to perform a continuous bottom-up clustering process by applying alternative rules for merging clusters. The outcome of this clustering process is a classification of the input proteins into a hierarchy of clusters of varying degrees of granularity. ProtoNet (version 1.3) is accessible in the form of an interactive web site at http://www.protonet.cs.huji.ac.il. ProtoNet provides navigation tools for monitoring the clustering process with a vertical and horizontal view. Each cluster at any level of the hierarchy is assigned with a statistical index, indicating the level of purity based on biological keywords such as those provided by SWISS-PROT and InterPro. ProtoNet can be used for function prediction, for defining superfamilies and subfamilies and for large-scale protein annotation purposes.
Assuntos
Bases de Dados de Proteínas , Proteínas/classificação , Animais , Análise por Conglomerados , Armazenamento e Recuperação da Informação , Internet , Proteínas/química , Proteínas/fisiologiaRESUMO
Sequence similarity is probably the most widely used tool to infer functional linkage between proteins. The fully sequenced, much researched, genome of Saccharomyces cerevisiae gives us on opportunity to compare and statistically quantify computational methods based on sequence similarity, which aim to detect such linkage. In addition, the amount of data regarding Saccharomyces Cerevisiae genes and proteins, which is not directly based on sequence is rapidly increasing. Consequently, it allows investigation of the connections and correlation between classification based on these types of data and that based solely on sequence similarity. In this work we start with a simple clustering algorithm to cluster genes based on the BLAST E-score of their similarity. We analyze how well one can infer function from these clusters and for how many of the genes that are currently unknown one can suggest a prediction. Given these parameters, we show that even a simple algorithm achieves better results than simply considering the BLAST output of matching genes. In the second part of the paper, we show that there is a highly significant correlation (p-value < 10(-4) for the vast majority of the experiments) between the aforementioned clusters and other types of classifications. Namely, we show that a pair of genes being clustered together is correlated with these genes having similar expression patterns in DNA array experiments and with the encoded proteins being involved in protein-protein interactions. Although this correlation is highly significant, it is, of course, not strong enough to be, by itself, a tool for predicting co-regulation of genes or interaction of proteins. We discuss possible explanations for this correlation. Furthermore, the statistical evaluation of these results should be considered when developing tools that are aimed at making such predictions.
Assuntos
Genes Fúngicos , Família Multigênica , Saccharomyces cerevisiae/genética , Algoritmos , Análise por Conglomerados , Biologia Computacional , DNA Fúngico/genética , Bases de Dados Genéticas , Perfilação da Expressão Gênica/estatística & dados numéricos , Ligação Genética , Análise de Sequência com Séries de Oligonucleotídeos/estatística & dados numéricos , Fases de Leitura Aberta , Proteínas de Saccharomyces cerevisiae/genéticaRESUMO
BACKGROUND: Transcription factors regulate gene expression by binding to specific cis-regulatory elements in gene promoters. Although DNA sequences that serve as transcription-factor binding sites have been characterized and associated with the regulation of numerous genes, the principles that govern the design and evolution of such sites are poorly understood. RESULTS: Using the comprehensive mapping of binding-site locations available in Saccharomyces cerevisiae, we examined possible factors that may have an impact on binding-site design. We found that binding sites tend to be shorter and fuzzier when they appear in promoter regions that bind multiple transcription factors. We further found that essential genes bind relatively fewer transcription factors, as do divergent promoters. We provide evidence that novel binding sites tend to appear in specific promoters that are already associated with multiple sites. CONCLUSION: Two principal models may account for the observed correlations. First, it may be that the interaction between multiple factors compensates for the decreased specificity of each specific binding sequence. In such a scenario, binding-site fuzziness is a consequence of the presence of multiple binding sites. Second, binding sites may tend to appear in promoter regions that are subject to low selective pressure, which also allows for fuzzier motifs. The latter possibility may account for the relatively low number of binding sites found in promoters of essential genes and in divergent promoters.