The spatial structure of the tumor immune microenvironment can explain and predict patient response in high-grade serous carcinoma.

Despite ovarian cancer being the deadliest gynecological malignancy, there has been little change to therapeutic options and mortality rates over the last three decades. Recent studies indicate that the composition of the tumor immune microenvironment (TIME) influences patient outcomes but are limited by a lack of spatial understanding. We performed multiplexed ion beam imaging (MIBI) on 83 human high-grade serous carcinoma tumors - one of the largest protein-based, spatially-intact, single-cell resolution tumor datasets assembled - and used statistical and machine learning approaches to connect features of the TIME spatial organization to patient outcomes. Along with traditional clinical/immunohistochemical attributes and indicators of TIME composition, we found that several features of TIME spatial organization had significant univariate correlations and/or high relative importance in high-dimensional predictive models. The top performing predictive model for patient progression-free survival (PFS) used a combination of TIME composition and spatial features. Results demonstrate the importance of spatial structure in understanding how the TIME contributes to treatment outcomes. Furthermore, the present study provides a generalizable roadmap for spatial analyses of the TIME in ovarian cancer research.

Sequential stacking link prediction algorithms for temporal networks.

Link prediction algorithms are indispensable tools in many scientific applications by speeding up network data collection and imputing missing connections. However, in many systems, links change over time and it remains unclear how to optimally exploit such temporal information for link predictions in such networks. Here, we show that many temporal topological features, in addition to having high computational cost, are less accurate in temporal link prediction than sequentially stacked static network features. This sequential stacking link prediction method uses 41 static network features that avoid detailed feature engineering choices and is capable of learning a highly accurate predictive distribution of future connections from historical data. We demonstrate that this algorithm works well for both partially observed and completely unobserved target layers, and on two temporal stochastic block models achieves near-oracle-level performance when combined with other single predictor methods as an ensemble learning method. Finally, we empirically illustrate that stacking multiple predictive methods together further improves performance on 19 real-world temporal networks from different domains.

Early Ovarian Cancer Detection in the Age of Fallopian Tube Precursors: A Systematic Review.

OBJECTIVE: To determine biomarkers other than CA 125 that could be used in identifying early-stage ovarian cancer. DATA SOURCES: Ovid MEDLINE ALL, EMBASE, Web of Science Core Collection, ScienceDirect, Clinicaltrials.gov , and CAB Direct were searched for English-language studies between January 2008 and April 2023 for the concepts of high-grade serous ovarian cancer, testing, and prevention or early diagnosis. METHODS OF STUDY SELECTION: The 5,523 related articles were uploaded to Covidence. Screening by two independent reviewers of the article abstracts led to the identification of 245 peer-reviewed primary research articles for full-text review. Full-text review by those reviewers led to the identification of 131 peer-reviewed primary research articles used for this review. TABULATION, INTEGRATION, AND RESULTS: Of 131 studies, only 55 reported sensitivity, specificity, or area under the curve (AUC), with 36 of the studies reporting at least one biomarker with a specificity of 80% or greater specificity or 0.9 or greater AUC. CONCLUSION: These findings suggest that although many types of biomarkers are being tested in ovarian cancer, most have similar or worse detection rates compared with CA 125 and have the same limitations of poor detection rates in early-stage disease. However, 27.5% of articles (36/131) reported biomarkers with better sensitivity and an AUC greater than 0.9 compared with CA 125 alone and deserve further exploration.

In Science Journals.

Highlights from the Science family of journals.

Gender and retention patterns among U.S. faculty.

Women remain underrepresented among faculty in nearly all academic fields. Using a census of 245,270 tenure-track and tenured professors at United States-based PhD-granting departments, we show that women leave academia overall at higher rates than men at every career age, in large part because of strongly gendered attrition at lower-prestige institutions, in non-STEM fields, and among tenured faculty. A large-scale survey of the same faculty indicates that the reasons faculty leave are gendered, even for institutions, fields, and career ages in which retention rates are not. Women are more likely than men to feel pushed from their jobs and less likely to feel pulled toward better opportunities, and women leave or consider leaving because of workplace climate more often than work-life balance. These results quantify the systemic nature of gendered faculty retention; contextualize its relationship with career age, institutional prestige, and field; and highlight the importance of understanding the gendered reasons for attrition rather than focusing on rates alone.

Labor advantages drive the greater productivity of faculty at elite universities.

Faculty at prestigious institutions dominate scientific discourse, producing a disproportionate share of all research publications. Environmental prestige can drive such epistemic disparity, but the mechanisms by which it causes increased faculty productivity remain unknown. Here, we combine employment, publication, and federal survey data for 78,802 tenure-track faculty at 262 PhD-granting institutions in the American university system to show through multiple lines of evidence that the greater availability of funded graduate and postdoctoral labor at more prestigious institutions drives the environmental effect of prestige on productivity. In particular, greater environmental prestige leads to larger faculty-led research groups, which drive higher faculty productivity, primarily in disciplines with group collaboration norms. In contrast, productivity does not increase substantially with prestige for faculty publications without group members or for group members themselves. The disproportionate scientific productivity of elite researchers can be largely explained by their substantial labor advantage rather than inherent differences in talent.

Quantifying hierarchy and dynamics in US faculty hiring and retention.

Faculty hiring and retention determine the composition of the US academic workforce and directly shape educational outcomes1, careers2, the development and spread of ideas3 and research priorities4,5. However, hiring and retention are dynamic, reflecting societal and academic priorities, generational turnover and efforts to diversify the professoriate along gender6-8, racial9 and socioeconomic10 lines. A comprehensive study of the structure and dynamics of the US professoriate would elucidate the effects of these efforts and the processes that shape scholarship more broadly. Here we analyse the academic employment and doctoral education of tenure-track faculty at all PhD-granting US universities over the decade 2011-2020, quantifying stark inequalities in faculty production, prestige, retention and gender. Our analyses show universal inequalities in which a small minority of universities supply a large majority of faculty across fields, exacerbated by patterns of attrition and reflecting steep hierarchies of prestige. We identify markedly higher attrition rates among faculty trained outside the United States or employed by their doctoral university. Our results indicate that gains in women's representation over this decade result from demographic turnover and earlier changes made to hiring, and are unlikely to lead to long-term gender parity in most fields. These analyses quantify the dynamics of US faculty hiring and retention, and will support efforts to improve the organization, composition and scholarship of the US academic workforce.

Socioeconomic roots of academic faculty.

Despite the special role of tenure-track faculty in society, training future researchers and producing scholarship that drives scientific and technological innovation, the sociodemographic characteristics of the professoriate have never been representative of the general population. Here we systematically investigate the indicators of faculty childhood socioeconomic status and consider how they may limit efforts to diversify the professoriate. Combining national-level data on education, income and university rankings with a 2017-2020 survey of 7,204 US-based tenure-track faculty across eight disciplines in STEM, social science and the humanities, we show that faculty are up to 25 times more likely to have a parent with a Ph.D. Moreover, this rate nearly doubles at prestigious universities and is stable across the past 50 years. Our results suggest that the professoriate is, and has remained, accessible disproportionately to the socioeconomically privileged, which is likely to deeply shape their scholarship and their reproduction.

Untangling the network effects of productivity and prominence among scientists.

While inequalities in science are common, most efforts to understand them treat scientists as isolated individuals, ignoring the network effects of collaboration. Here, we develop models that untangle the network effects of productivity defined as paper counts, and prominence referring to high-impact publications, of individual scientists from their collaboration networks. We find that gendered differences in the productivity and prominence of mid-career researchers can be largely explained by differences in their coauthorship networks. Hence, collaboration networks act as a form of social capital, and we find evidence of their transferability from senior to junior collaborators, with benefits that decay as researchers age. Collaboration network effects can also explain a large proportion of the productivity and prominence advantages held by researchers at prestigious institutions. These results highlight a substantial role of social networks in driving inequalities in science, and suggest that collaboration networks represent an important form of unequally distributed social capital that shapes who makes what scientific discoveries.

In Science Journals.

[Figure: see text].

In Science Journals.

Highlights from the Science family of journals.

Examining the consumption of radical content on YouTube.

Although it is under-studied relative to other social media platforms, YouTube is arguably the largest and most engaging online media consumption platform in the world. Recently, YouTube's scale has fueled concerns that YouTube users are being radicalized via a combination of biased recommendations and ostensibly apolitical "anti-woke" channels, both of which have been claimed to direct attention to radical political content. Here we test this hypothesis using a representative panel of more than 300,000 Americans and their individual-level browsing behavior, on and off YouTube, from January 2016 through December 2019. Using a labeled set of political news channels, we find that news consumption on YouTube is dominated by mainstream and largely centrist sources. Consumers of far-right content, while more engaged than average, represent a small and stable percentage of news consumers. However, consumption of "anti-woke" content, defined in terms of its opposition to progressive intellectual and political agendas, grew steadily in popularity and is correlated with consumption of far-right content off-platform. We find no evidence that engagement with far-right content is caused by YouTube recommendations systematically, nor do we find clear evidence that anti-woke channels serve as a gateway to the far right. Rather, consumption of political content on YouTube appears to reflect individual preferences that extend across the web as a whole.

Decoding the dynamic tumor microenvironment.

An opportunity to improve cancer outcomes with machine learning.

Denoising large-scale biological data using network filters.

BACKGROUND: Large-scale biological data sets are often contaminated by noise, which can impede accurate inferences about underlying processes. Such measurement noise can arise from endogenous biological factors like cell cycle and life history variation, and from exogenous technical factors like sample preparation and instrument variation. RESULTS: We describe a general method for automatically reducing noise in large-scale biological data sets. This method uses an interaction network to identify groups of correlated or anti-correlated measurements that can be combined or "filtered" to better recover an underlying biological signal. Similar to the process of denoising an image, a single network filter may be applied to an entire system, or the system may be first decomposed into distinct modules and a different filter applied to each. Applied to synthetic data with known network structure and signal, network filters accurately reduce noise across a wide range of noise levels and structures. Applied to a machine learning task of predicting changes in human protein expression in healthy and cancerous tissues, network filtering prior to training increases accuracy up to 43% compared to using unfiltered data. CONCLUSIONS: Network filters are a general way to denoise biological data and can account for both correlation and anti-correlation between different measurements. Furthermore, we find that partitioning a network prior to filtering can significantly reduce errors in networks with heterogenous data and correlation patterns, and this approach outperforms existing diffusion based methods. Our results on proteomics data indicate the broad potential utility of network filters to applications in systems biology.

The unequal impact of parenthood in academia.

Across academia, men and women tend to publish at unequal rates. Existing explanations include the potentially unequal impact of parenthood on scholarship, but a lack of appropriate data has prevented its clear assessment. Here, we quantify the impact of parenthood on scholarship using an extensive survey of the timing of parenthood events, longitudinal publication data, and perceptions of research expectations among 3064 tenure-track faculty at 450 Ph.D.-granting computer science, history, and business departments across the United States and Canada, along with data on institution-specific parental leave policies. Parenthood explains most of the gender productivity gap by lowering the average short-term productivity of mothers, even as parents tend to be slightly more productive on average than nonparents. However, the size of productivity penalty for mothers appears to have shrunk over time. Women report that paid parental leave and adequate childcare are important factors in their recruitment and retention. These results have broad implications for efforts to improve the inclusiveness of scholarship.

The Capacity of the Ovarian Cancer Tumor Microenvironment to Integrate Inflammation Signaling Conveys a Shorter Disease-free Interval.

PURPOSE: Ovarian cancer has one of the highest deaths to incidence ratios across all cancers. Initial chemotherapy is effective, but most patients develop chemoresistant disease. Mechanisms driving clinical chemo-response or -resistance are not well-understood. However, achieving optimal surgical cytoreduction improves survival, and cytoreduction is improved by neoadjuvant chemotherapy (NACT). NACT offers a window to profile pre- versus post-NACT tumors, which we used to identify chemotherapy-induced changes to the tumor microenvironment. EXPERIMENTAL DESIGN: We obtained matched pre- and post-NACT archival tumor tissues from patients with high-grade serous ovarian cancer (patient, n = 6). We measured mRNA levels of 770 genes (756 genes/14 housekeeping genes, NanoString Technologies), and performed reverse phase protein array (RPPA) on a subset of matched tumors. We examined cytokine levels in pre-NACT ascites samples (n = 39) by ELISAs. A tissue microarray with 128 annotated ovarian tumors expanded the transcriptional, RPPA, and cytokine data by multispectral IHC. RESULTS: The most upregulated gene post-NACT was IL6 (16.79-fold). RPPA data were concordant with mRNA, consistent with elevated immune infiltration. Elevated IL6 in pre-NACT ascites specimens correlated with a shorter time to recurrence. Integrating NanoString (n = 12), RPPA (n = 4), and cytokine (n = 39) studies identified an activated inflammatory signaling network and induced IL6 and IER3 (immediate early response 3) post-NACT, associated with poor chemo-response and time to recurrence. CONCLUSIONS: Multiomics profiling of ovarian tumor samples pre- and post-NACT provides unique insight into chemo-induced changes to the tumor microenvironment. We identified a novel IL6/IER3 signaling axis that may drive chemoresistance and disease recurrence.

Stacking models for nearly optimal link prediction in complex networks.

Most real-world networks are incompletely observed. Algorithms that can accurately predict which links are missing can dramatically speed up network data collection and improve network model validation. Many algorithms now exist for predicting missing links, given a partially observed network, but it has remained unknown whether a single best predictor exists, how link predictability varies across methods and networks from different domains, and how close to optimality current methods are. We answer these questions by systematically evaluating 203 individual link predictor algorithms, representing three popular families of methods, applied to a large corpus of 550 structurally diverse networks from six scientific domains. We first show that individual algorithms exhibit a broad diversity of prediction errors, such that no one predictor or family is best, or worst, across all realistic inputs. We then exploit this diversity using network-based metalearning to construct a series of "stacked" models that combine predictors into a single algorithm. Applied to a broad range of synthetic networks, for which we may analytically calculate optimal performance, these stacked models achieve optimal or nearly optimal levels of accuracy. Applied to real-world networks, stacked models are superior, but their accuracy varies strongly by domain, suggesting that link prediction may be fundamentally easier in social networks than in biological or technological networks. These results indicate that the state of the art for link prediction comes from combining individual algorithms, which can achieve nearly optimal predictions. We close with a brief discussion of limitations and opportunities for further improvements.