Pub Date : 2024-09-24DOI: 10.1038/s43588-024-00695-4
Sophia Chen
Many AI companies implement safety systems to protect users from offensive or inaccurate content. Though well intentioned, these filters can exacerbate existing inequalities, and data shows that they have disproportionately removed LGBTQ+ content.
{"title":"The lost data: how AI systems censor LGBTQ+ content in the name of safety","authors":"Sophia Chen","doi":"10.1038/s43588-024-00695-4","DOIUrl":"10.1038/s43588-024-00695-4","url":null,"abstract":"Many AI companies implement safety systems to protect users from offensive or inaccurate content. Though well intentioned, these filters can exacerbate existing inequalities, and data shows that they have disproportionately removed LGBTQ+ content.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"629-632"},"PeriodicalIF":12.0,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00695-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142317007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-24DOI: 10.1038/s43588-024-00686-5
Joaquín Barroso-Flores
The current global economy heavily relies on digital and data-based technologies, which have the use of supercomputing at their core. Latin America is a vast source of human talent in computer science, but the lag in infrastructure investment due to economic and political struggles may cause the economic development of the region to fall behind.
{"title":"Accelerating economic development in Latin America through overcoming access challenges to supercomputing infrastructure","authors":"Joaquín Barroso-Flores","doi":"10.1038/s43588-024-00686-5","DOIUrl":"10.1038/s43588-024-00686-5","url":null,"abstract":"The current global economy heavily relies on digital and data-based technologies, which have the use of supercomputing at their core. Latin America is a vast source of human talent in computer science, but the lag in infrastructure investment due to economic and political struggles may cause the economic development of the region to fall behind.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"644-645"},"PeriodicalIF":12.0,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00686-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142317008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-24DOI: 10.1038/s43588-024-00702-8
We present a Focus that calls attention to the current state of diversity, equity, and inclusion in computational science, including discussions on the challenges of improving equitable access and representation, as well as on strategies for improving computational tools to avoid contributing to inequalities.
{"title":"Putting a spotlight on diversity, equity, and inclusion","authors":"","doi":"10.1038/s43588-024-00702-8","DOIUrl":"10.1038/s43588-024-00702-8","url":null,"abstract":"We present a Focus that calls attention to the current state of diversity, equity, and inclusion in computational science, including discussions on the challenges of improving equitable access and representation, as well as on strategies for improving computational tools to avoid contributing to inequalities.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"627-628"},"PeriodicalIF":12.0,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00702-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142316996","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-24DOI: 10.1038/s43588-024-00694-5
Siddharth Suri
We can design, build and use AI systems with intentionality, to make them an equalizing force within society, or we can use AI without intentionality, in which case AI could become a force that exacerbates inequality, or both. Society has the power to decide which.
{"title":"Defining our future with generative AI","authors":"Siddharth Suri","doi":"10.1038/s43588-024-00694-5","DOIUrl":"10.1038/s43588-024-00694-5","url":null,"abstract":"We can design, build and use AI systems with intentionality, to make them an equalizing force within society, or we can use AI without intentionality, in which case AI could become a force that exacerbates inequality, or both. Society has the power to decide which.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"641-643"},"PeriodicalIF":12.0,"publicationDate":"2024-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s43588-024-00694-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142316997","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-23DOI: 10.1038/s43588-024-00696-3
We developed mixture model inference with discrete-coupled autoencoders (MMIDAS), an unsupervised variational framework that jointly learns discrete clusters and continuous cluster-specific variability. When applied to unimodal or multimodal single-cell omic data, MMIDAS learned single-cell representations with robust cell type definitions and interpretable, continuous within-cell type variability.
{"title":"Joint inference of discrete and continuous factors captures variability across and within cell types","authors":"","doi":"10.1038/s43588-024-00696-3","DOIUrl":"10.1038/s43588-024-00696-3","url":null,"abstract":"We developed mixture model inference with discrete-coupled autoencoders (MMIDAS), an unsupervised variational framework that jointly learns discrete clusters and continuous cluster-specific variability. When applied to unimodal or multimodal single-cell omic data, MMIDAS learned single-cell representations with robust cell type definitions and interpretable, continuous within-cell type variability.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 10","pages":"733-734"},"PeriodicalIF":12.0,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142309285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Reproducible definition and identification of cell types is essential to enable investigations into their biological function and to understand their relevance in the context of development, disease and evolution. Current approaches model variability in data as continuous latent factors, followed by clustering as a separate step, or immediately apply clustering on the data. We show that such approaches can suffer from qualitative mistakes in identifying cell types robustly, particularly when the number of such cell types is in the hundreds or even thousands. Here we propose an unsupervised method, Mixture Model Inference with Discrete-coupled AutoencoderS (MMIDAS), which combines a generalized mixture model with a multi-armed deep neural network to jointly infer the discrete type and continuous type-specific variability. Using four recent datasets of brain cells spanning different technologies, species and conditions, we demonstrate that MMIDAS can identify reproducible cell types and infer cell type-dependent continuous variability in both unimodal and multimodal datasets. Clustering in high-dimensional spaces with a large number of clusters and identifying common aspects of within-cluster variability remain challenging. Here the authors develop an unsupervised method for this purpose and demonstrate it on brain single-cell datasets.
{"title":"Joint inference of discrete cell types and continuous type-specific variability in single-cell datasets with MMIDAS","authors":"Yeganeh Marghi, Rohan Gala, Fahimeh Baftizadeh, Uygar Sümbül","doi":"10.1038/s43588-024-00683-8","DOIUrl":"10.1038/s43588-024-00683-8","url":null,"abstract":"Reproducible definition and identification of cell types is essential to enable investigations into their biological function and to understand their relevance in the context of development, disease and evolution. Current approaches model variability in data as continuous latent factors, followed by clustering as a separate step, or immediately apply clustering on the data. We show that such approaches can suffer from qualitative mistakes in identifying cell types robustly, particularly when the number of such cell types is in the hundreds or even thousands. Here we propose an unsupervised method, Mixture Model Inference with Discrete-coupled AutoencoderS (MMIDAS), which combines a generalized mixture model with a multi-armed deep neural network to jointly infer the discrete type and continuous type-specific variability. Using four recent datasets of brain cells spanning different technologies, species and conditions, we demonstrate that MMIDAS can identify reproducible cell types and infer cell type-dependent continuous variability in both unimodal and multimodal datasets. Clustering in high-dimensional spaces with a large number of clusters and identifying common aspects of within-cluster variability remain challenging. Here the authors develop an unsupervised method for this purpose and demonstrate it on brain single-cell datasets.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"706-722"},"PeriodicalIF":12.0,"publicationDate":"2024-09-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142317001","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-20DOI: 10.1038/s43588-024-00689-2
Tara Chari, Gennady Gorin, Lior Pachter
Multimodal, single-cell genomics technologies enable simultaneous measurement of multiple facets of DNA and RNA processing in the cell. This creates opportunities for transcriptome-wide, mechanistic studies of cellular processing in heterogeneous cell populations, such as regulation of cell fate by transcriptional stochasticity or tumor proliferation through aberrant splicing dynamics. However, current methods for determining cell types or ‘clusters’ in multimodal data often rely on ad hoc approaches to balance or integrate measurements, and assumptions ignoring inherent properties of the data. To enable interpretable and consistent cell cluster determination, we present meK-means (mechanistic K-means) which integrates modalities through a unifying model of transcription to learn underlying, shared biophysical states. With meK-means we can cluster cells with nascent and mature mRNA measurements, utilizing the causal, physical relationships between these modalities. This identifies shared transcription dynamics across cells, which induce the observed molecule counts, and provides an alternative definition for ‘clusters’ through the governing parameters of cellular processes. MeK-means clusters single-cell multimodal data by linking modalities through their biophysical relationships. We redefine clusters through transcription kinetics to reveal how RNA production and processing drive cellular diversity and disease.
{"title":"Biophysically interpretable inference of cell types from multimodal sequencing data","authors":"Tara Chari, Gennady Gorin, Lior Pachter","doi":"10.1038/s43588-024-00689-2","DOIUrl":"10.1038/s43588-024-00689-2","url":null,"abstract":"Multimodal, single-cell genomics technologies enable simultaneous measurement of multiple facets of DNA and RNA processing in the cell. This creates opportunities for transcriptome-wide, mechanistic studies of cellular processing in heterogeneous cell populations, such as regulation of cell fate by transcriptional stochasticity or tumor proliferation through aberrant splicing dynamics. However, current methods for determining cell types or ‘clusters’ in multimodal data often rely on ad hoc approaches to balance or integrate measurements, and assumptions ignoring inherent properties of the data. To enable interpretable and consistent cell cluster determination, we present meK-means (mechanistic K-means) which integrates modalities through a unifying model of transcription to learn underlying, shared biophysical states. With meK-means we can cluster cells with nascent and mature mRNA measurements, utilizing the causal, physical relationships between these modalities. This identifies shared transcription dynamics across cells, which induce the observed molecule counts, and provides an alternative definition for ‘clusters’ through the governing parameters of cellular processes. MeK-means clusters single-cell multimodal data by linking modalities through their biophysical relationships. We redefine clusters through transcription kinetics to reveal how RNA production and processing drive cellular diversity and disease.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"677-689"},"PeriodicalIF":12.0,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142317000","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-20DOI: 10.1038/s43588-024-00691-8
Yicheng Gao, Qi Liu
A recent study proposes an approach that integrates unspliced and spliced mRNA count data by leveraging shared biophysical states across cells, offering a more interpretable and consistent framework for determining cell clusters based on transcriptional kinetics.
{"title":"Delineating cell types with transcriptional kinetics","authors":"Yicheng Gao, Qi Liu","doi":"10.1038/s43588-024-00691-8","DOIUrl":"10.1038/s43588-024-00691-8","url":null,"abstract":"A recent study proposes an approach that integrates unspliced and spliced mRNA count data by leveraging shared biophysical states across cells, offering a more interpretable and consistent framework for determining cell clusters based on transcriptional kinetics.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"657-658"},"PeriodicalIF":12.0,"publicationDate":"2024-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142316999","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-16DOI: 10.1038/s43588-024-00688-3
Shenghao Wu, Chengcheng Huang, Adam C. Snyder, Matthew A. Smith, Brent Doiron, Byron M. Yu
Understanding brain function is facilitated by constructing computational models that accurately reproduce aspects of brain activity. Networks of spiking neurons capture the underlying biophysics of neuronal circuits, yet their activity’s dependence on model parameters is notoriously complex. As a result, heuristic methods have been used to configure spiking network models, which can lead to an inability to discover activity regimes complex enough to match large-scale neuronal recordings. Here we propose an automatic procedure, Spiking Network Optimization using Population Statistics (SNOPS), to customize spiking network models that reproduce the population-wide covariability of large-scale neuronal recordings. We first confirmed that SNOPS accurately recovers simulated neural activity statistics. Then, we applied SNOPS to recordings in macaque visual and prefrontal cortices and discovered previously unknown limitations of spiking network models. Taken together, SNOPS can guide the development of network models, thereby enabling deeper insight into how networks of neurons give rise to brain function. An automatic framework, SNOPS, is developed for configuring a spiking network model to reproduce neuronal recordings. It is used to discover previously unknown limitations of spiking network models, thereby guiding model development.
{"title":"Automated customization of large-scale spiking network models to neuronal population activity","authors":"Shenghao Wu, Chengcheng Huang, Adam C. Snyder, Matthew A. Smith, Brent Doiron, Byron M. Yu","doi":"10.1038/s43588-024-00688-3","DOIUrl":"10.1038/s43588-024-00688-3","url":null,"abstract":"Understanding brain function is facilitated by constructing computational models that accurately reproduce aspects of brain activity. Networks of spiking neurons capture the underlying biophysics of neuronal circuits, yet their activity’s dependence on model parameters is notoriously complex. As a result, heuristic methods have been used to configure spiking network models, which can lead to an inability to discover activity regimes complex enough to match large-scale neuronal recordings. Here we propose an automatic procedure, Spiking Network Optimization using Population Statistics (SNOPS), to customize spiking network models that reproduce the population-wide covariability of large-scale neuronal recordings. We first confirmed that SNOPS accurately recovers simulated neural activity statistics. Then, we applied SNOPS to recordings in macaque visual and prefrontal cortices and discovered previously unknown limitations of spiking network models. Taken together, SNOPS can guide the development of network models, thereby enabling deeper insight into how networks of neurons give rise to brain function. An automatic framework, SNOPS, is developed for configuring a spiking network model to reproduce neuronal recordings. It is used to discover previously unknown limitations of spiking network models, thereby guiding model development.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"690-705"},"PeriodicalIF":12.0,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142255499","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-09-12DOI: 10.1038/s43588-024-00690-9
Jie Hu
A computational model is proposed to provide a better understanding of human altruism, highlighting the role of multiple motives that influence altruistic behaviors.
为了更好地理解人类的利他主义，我们提出了一个计算模型，强调影响利他行为的多种动机的作用。
{"title":"Deconstructing the compounds of altruism","authors":"Jie Hu","doi":"10.1038/s43588-024-00690-9","DOIUrl":"10.1038/s43588-024-00690-9","url":null,"abstract":"A computational model is proposed to provide a better understanding of human altruism, highlighting the role of multiple motives that influence altruistic behaviors.","PeriodicalId":74246,"journal":{"name":"Nature computational science","volume":"4 9","pages":"655-656"},"PeriodicalIF":12.0,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142208697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：