Explore the vast range of titles available.

Optimizing molecular design across expansive chemical spaces presents unique challenges, especially in maintaining predictive accuracy under domain shifts. This study integrates uncertainty quantification (UQ), directed message passing neural networks (D-MPNNs), and genetic algorithms (GAs) to address these challenges. We systematically evaluate whether UQ-enhanced D-MPNNs can effectively optimize broad, open-ended chemical spaces and identify the most effective implementation strategies. Using benchmarks from the Tartarus and GuacaMol platforms, our results show that UQ integration via probabilistic improvement optimization (PIO) enhances optimization success in most cases, supporting more reliable exploration of chemically diverse regions. In multi-objective tasks, PIO proves especially advantageous, balancing competing objectives and outperforming uncertainty-agnostic approaches. This work provides practical guidelines for integrating UQ in computational-aided molecular design (CAMD). Optimizing molecular design across chemical spaces is challenging. Here, authors integrate uncertainty quantification with graph neural networks and genetic algorithms, demonstrating that probabilistic improvement optimization enhances success rates in molecular discovery.

Posttraumatic stress disorder (PTSD) significantly impacts mental health and society, yet gaps exist in understanding its temporal relationships with comorbid mental disorders. This study used self-comparison approach to explore the longitudinal relationships between PTSD and comorbid mental disorders, suicide deaths, and all cause of death, identifying risk periods and temporal sequences. A cohort study using the Taiwan National Health Insurance Research Database included adults with incident PTSD diagnosis from 2000 to 2012, followed until 2018 or death. The study employed a self-controlled case series design, and risk period encompasses the present year and five years both preceding and following the incident PTSD diagnosis, with time outside the risk period functioned as a self-controlled period for comparison. Analysis showed reciprocal relationships between PTSD and various mental disorders, especially within the year of or preceding PTSD diagnosis. However, major depressive disorders primarily emerged as a predictor of PTSD. Furthermore, an increased risk of suicide deaths was noted within three to five years post-PTSD, while no association with all cause of death was observed within five years post-PTSD. This study reveals complex, bidirectional relationships between PTSD and several mental disorders, emphasizing the need for integrated treatment and targeted prevention. Immediate attention to PTSD and comorbid mental disorders is crucial, particularly within the first year. The findings provide deeper insights for psychiatric nosology, psychiatric disorder diagnosis, treatment, and prevention.

Ukraine has been embroiled in an increasing war since February 2022. In addition to Ukrainians, the Russo-Ukraine war has affected Poles due to the refugee crisis and the Taiwanese, who are facing a potential crisis with China. We examined the mental health status and associated factors in Ukraine, Poland, and Taiwan. The data will be used for future reference as the war is still ongoing. From March 8 to April 26, 2022, we conducted an online survey using snowball sampling techniques in Ukraine, Poland, and Taiwan. Depression, anxiety, and stress were measured using the Depression, Anxiety, and Stress (DASS)-21 item scale; post-traumatic stress symptoms by the Impact of Event Scale-Revised (IES-R) and coping strategies by the Coping Orientation to Problems Experienced Inventory (Brief-COPE). We used multivariate linear regression to identify factors significantly associated with DASS-21 and IES-R scores. There were 1626 participants (Poland: 1053; Ukraine: 385; Taiwan: 188) in this study. Ukrainian participants reported significantly higher DASS-21 ( p  < 0.001) and IES-R ( p  < 0.01) scores than Poles and Taiwanese. Although Taiwanese participants were not directly involved in the war, their mean IES-R scores (40.37 ± 16.86) were only slightly lower than Ukrainian participants (41.36 ± 14.94). Taiwanese reported significantly higher avoidance scores (1.60 ± 0.47) than the Polish (0.87 ± 0.53) and Ukrainian (0.91 ± 0.5) participants ( p  < 0.001). More than half of the Taiwanese (54.3%) and Polish (80.3%) participants were distressed by the war scenes in the media. More than half (52.5%) of the Ukrainian participants would not seek psychological help despite a significantly higher prevalence of psychological distress. Multivariate linear regression analyses found that female gender, Ukrainian and Polish citizenship, household size, self-rating health status, past psychiatric history, and avoidance coping were significantly associated with higher DASS-21 and IES-R scores after adjustment of other variables ( p  < 0.05). We have identified mental health sequelae in Ukrainian, Poles, and Taiwanese with the ongoing Russo-Ukraine war. Risk factors associated with developing depression, anxiety, stress, and post-traumatic stress symptoms include female gender, self-rating health status, past psychiatric history, and avoidance coping. Early resolution of the conflict, online mental health interventions, delivery of psychotropic medications, and distraction techniques may help to improve the mental health of people who stay inside and outside Ukraine.

This review surveys the recent advances and challenges in predicting and optimizing reaction conditions using machine learning techniques. The paper emphasizes the importance of acquiring and processing large and diverse datasets of chemical reactions, and the use of both global and local models to guide the design of synthetic processes. Global models exploit the information from comprehensive databases to suggest general reaction conditions for new reactions, while local models fine-tune the specific parameters for a given reaction family to improve yield and selectivity. The paper also identifies the current limitations and opportunities in this field, such as the data quality and availability, and the integration of high-throughput experimentation. The paper demonstrates how the combination of chemical engineering, data science, and ML algorithms can enhance the efficiency and effectiveness of reaction conditions design, and enable novel discoveries in synthetic chemistry.

In the field of chemical synthesis planning, the accurate recommendation of reaction conditions is essential for achieving successful outcomes. This work introduces an innovative deep learning approach designed to address the complex task of predicting appropriate reagents, solvents, and reaction temperatures for chemical reactions. Our proposed methodology combines a multi-label classification model with a ranking model to offer tailored reaction condition recommendations based on relevance scores derived from anticipated product yields. To tackle the challenge of limited data for unfavorable reaction contexts, we employed the technique of hard negative sampling to generate reaction conditions that might be mistakenly classified as suitable, forcing the model to refine its decision boundaries, especially in challenging cases. Our developed model excels in proposing conditions where an exact match to the recorded solvents and reagents is found within the top-10 predictions 73% of the time. It also predicts temperatures within ± 20 ° C of the recorded temperature in 89% of test cases. Notably, the model demonstrates its capacity to recommend multiple viable reaction conditions, with accuracy varying based on the availability of condition records associated with each reaction. What sets this model apart is its ability to suggest alternative reaction conditions beyond the constraints of the dataset. This underscores its potential to inspire innovative approaches in chemical research, presenting a compelling opportunity for advancing chemical synthesis planning and elevating the field of reaction engineering. Scientific contribution The combination of multi-label classification and ranking models provides tailored recommendations for reaction conditions based on the reaction yields. A novel approach is presented to address the issue of data scarcity in negative reaction conditions through data augmentation. Graphical Abstract

The suicide risk on Chinese Valentine's Day and its potential risk factors have not been examined. This study assessed whether the suicide rates around Chinese and Western Valentine's Days differed from the rest of the year in Taiwan, and the roles of various genders and marital statuses. This study analyzed daily suicide data from Taiwan's Cause of Death Statistics between January 2012 and December 2022. We compared the suicide rate of each day in the week before and after Chinese and Western Valentine's Days with those of the remainder of the year using Quasi-Poisson regression models stratified by gender and marital status. We then performed a moderation analysis to explore whether the effect of Valentine's Day on suicide differed by gender. Married women reported a higher suicide risk on the third day after Chinese Valentine's Day than married men did. Although a similar trend was observed in Western Valentine's Day between married women and men, it did not reach statistical significance. Suicide rates for certain days in the week before or after Chinese and Western Valentine's Days were different from other days of the year, and these differences were gender- and marital status-specific.

This paper presents AutoTemplate, an innovative data preprocessing protocol, addressing the crucial need for high-quality chemical reaction datasets in the realm of machine learning applications in organic chemistry. Recent advances in artificial intelligence have expanded the application of machine learning in chemistry, particularly in yield prediction, retrosynthesis, and reaction condition prediction. However, the effectiveness of these models hinges on the integrity of chemical reaction datasets, which are often plagued by inconsistencies like missing reactants, incorrect atom mappings, and outright erroneous reactions. AutoTemplate introduces a two-stage approach to refine these datasets. The first stage involves extracting meaningful reaction transformation rules and formulating generic reaction templates using a simplified SMARTS representation. This simplification broadens the applicability of templates across various chemical reactions. The second stage is template-guided reaction curation, where these templates are systematically applied to validate and correct the reaction data. This process effectively amends missing reactant information, rectifies atom-mapping errors, and eliminates incorrect data entries. A standout feature of AutoTemplate is its capability to concurrently identify and correct false chemical reactions. It operates on the premise that most reactions in datasets are accurate, using these as templates to guide the correction of flawed entries. The protocol demonstrates its efficacy across a range of chemical reactions, significantly enhancing dataset quality. This advancement provides a more robust foundation for developing reliable machine learning models in chemistry, thereby improving the accuracy of forward and retrosynthetic predictions. AutoTemplate marks a significant progression in the preprocessing of chemical reaction datasets, bridging a vital gap and facilitating more precise and efficient machine learning applications in organic synthesis. Scientific contribution The proposed automated preprocessing tool for chemical reaction data aims to identify errors within chemical databases. Specifically, if the errors involve atom mapping or the absence of reactant types, corrections can be systematically applied using reaction templates, ultimately elevating the overall quality of the database. Graphical Abstract

Evidence on sex-specific incidence and comorbidity risk factors of suicide among patients with bipolar disorder is scarce. This study investigated the sex-specific risk profiles for suicide among the bipolar disorder population in terms of incidence, healthcare utilization and comorbidity. Using data from the Taiwan National Health Insurance Research Database between 1 January 2000 and 31 December 2016, this nationwide cohort study included patients with bipolar disorder ( = 46 490) and individuals representative of the general population ( = 185 960) matched by age and sex at a 1:4 ratio. Mortality rate ratios (MRRs) of suicide were calculated between suicide rates of bipolar disorder cohort and general population. In addition, a nested case-control study (1428 cases died by suicide and 5710 living controls) was conducted in the bipolar disorder cohort to examine the sex-specific risk of healthcare utilization and comorbidities. Suicide risk was considerably higher in the cohort (MRR = 21.9) than in the general population, especially among women (MRR = 35.6). Sex-stratified analyses revealed distinct healthcare utilization patterns and physical comorbidity risk profiles between the sexes. Although female patients who died by suicide had higher risks of nonhypertensive cardiovascular disease, pneumonia, chronic kidney disease, peptic ulcer, irritable bowel syndrome, and sepsis compared to their living counterparts, male patients who died by suicide had higher risks of chronic kidney disease and sepsis compared to the living controls. Patients with bipolar disorder who died by suicide had sex-specific risk profiles in incidence and physical comorbidities. Identifying these modifiable risk factors may guide interventions for suicide risk reduction.

Coronavirus disease 2019 (COVID-19) pandemic has impacted many aspects of people’s lives all over the world. This Facebook survey study aimed to investigate the COVID-19-related factors that were associated with sleep disturbance and suicidal thoughts among members of the public during the COVID-19 pandemic in Taiwan. The online survey recruited 1970 participants through a Facebook advertisement. Their self-reported experience of sleep disturbance and suicidal thoughts in the previous week were collected along with a number of COVID-19-related factors, including level of worry, change in social interaction and daily lives, any academic/occupational interference, levels of social and specific support, and self-reported physical health. In total, 55.8% of the participants reported sleep disturbance, and 10.8% reported having suicidal thoughts in the previous week. Multiple COVID-19-related factors were associated with sleep disturbance and suicidal thoughts in the COVID-19 pandemic. Increased worry about COVID-19, more severe impact of COVID-19 on social interaction, lower perceived social support, more severe academic/occupational interference due to COVID-19, lower COVID-19-specified support, and poorer self-reported physical health were significantly associated with sleep disturbance. Less handwashing, lower perceived social support, lower COVID-19-specified support, poorer self-reported physical health, and younger age were significantly associated with suicidal thoughts. Further investigation is needed to understand the changes in mental health among the public since the mitigation of the COVID-19 pandemic.

There has been a lack of prevalence estimates of DSM-5 mental disorders in child populations at the national level worldwide. This study estimated the lifetime and 6-month prevalence of mental disorders according to the DSM-5 diagnostic criteria in Taiwanese children. Taiwan's National Epidemiological Study of Child Mental Disorders used the stratified cluster sampling to select 69 schools in Taiwan resulting in a nationally representative sample of 4816 children in grades 3 (n = 1352), 5 (n = 1297) and 7 (n = 2167). All the participants underwent face-to-face psychiatric interviews using the Kiddie-Schedule for Affective Disorders and Schizophrenia-Epidemiological version, modified for the DSM-5, and they and their parents completed questionnaires. The inverse probability censoring weighting (IPCW)-adjusted prevalence was reported to minimise non-response bias. The IPCW-adjusted prevalence rates of mental disorders decreased by 0.1-0.5% than raw weighted prevalence. The IPCW-adjusted weighted lifetime and 6-month prevalence rates for overall mental disorders were 31.6 and 25.0%, respectively. The most prevalent mental disorders (lifetime, 6-month) were anxiety disorders (15.2, 12.0%) and attention-deficit hyperactivity disorder (10.1, 8.7%), followed by sleep disorders, tic disorders, oppositional defiant disorder and autism spectrum disorder. The prevalence rates of new DSM-5 mental disorders, avoidant/restrictive food intake disorder and disruptive mood dysregulation disorder were low (<1%). Our findings, similar to the DSM-IV prevalence rates reported in Western countries, indicate that DSM-5 mental disorders are common in the Taiwanese child population and suggest the need for public awareness, early detection and prevention.