Cell-cell interactions orchestrate organismal development, homeostasis and single-cell functions. When cells do not properly interact or improperly decode molecular messages, disease ensues. Thus, the identification and quantification of intercellular signalling pathways has become a common analysis performed across diverse disciplines. The expansion of protein-protein interaction databases and recent advances in RNA sequencing technologies have enabled routine analyses of intercellular signalling from gene expression measurements of bulk and single-cell data sets. In particular, ligand-receptor pairs can be used to infer intercellular communication from the coordinated expression of their cognate genes. In this Review, we highlight discoveries enabled by analyses of cell-cell interactions from transcriptomic data and review the methods and tools used in this context.

/pdf/deciphering-cell-cell-interactions-and-communication-from-19hzux0gfi.pdf

Deciphering cell-cell interactions and communication from gene expression.

Objectives To investigate severe COVID-19 risk by occupational group. Methods Baseline UK Biobank data (2006–10) for England were linked to SARS-CoV-2 test results from Public Health England (16 March to 26 July 2020). Included participants were employed or self-employed at baseline, alive and aged Results Of 120 075 participants, 271 had severe COVID-19. Relative to non-essential workers, healthcare workers (RR 7.43, 95% CI 5.52 to 10.00), social and education workers (RR 1.84, 95% CI 1.21 to 2.82) and other essential workers (RR 1.60, 95% CI 1.05 to 2.45) had a higher risk of severe COVID-19. Using more detailed groupings, medical support staff (RR 8.70, 95% CI 4.87 to 15.55), social care (RR 2.46, 95% CI 1.47 to 4.14) and transport workers (RR 2.20, 95% CI 1.21 to 4.00) had the highest risk within the broader groups. Compared with white non-essential workers, non-white non-essential workers had a higher risk (RR 3.27, 95% CI 1.90 to 5.62) and non-white essential workers had the highest risk (RR 8.34, 95% CI 5.17 to 13.47). Using SOC 2000 major groups, associate professional and technical occupations, personal service occupations and plant and machine operatives had a higher risk, compared with managers and senior officials. Conclusions Essential workers have a higher risk of severe COVID-19. These findings underscore the need for national and organisational policies and practices that protect and support workers with an elevated risk of severe COVID-19.

https://oem.bmj.com/content/oemed/early/2020/12/01/oemed-2020-106731.full.pdf

Occupation and risk of severe COVID-19: prospective cohort study of 120 075 UK Biobank participants.

Background Estimating the risk of pre-existing comorbidities on coronavirus disease 2019 (COVID-19) mortality may promote the importance of targeting populations at risk to improve survival. This systematic review and meta-analysis aimed to estimate the association of pre-existing comorbidities with COVID-19 mortality. Methods We searched MEDLINE, SCOPUS, OVID, and Cochrane Library databases, and medrxiv.org from December 1st, 2019, to July 9th, 2020. The outcome of interest was the risk of COVID-19 mortality in patients with and without pre-existing comorbidities. We analyzed 11 comorbidities: cardiovascular diseases, hypertension, diabetes, congestive heart failure, cerebrovascular disease, chronic kidney disease, chronic liver disease, cancer, chronic obstructive pulmonary disease, asthma, and HIV/AIDS. Two reviewers independently extracted data and assessed the risk of bias. All analyses were performed using random-effects models and heterogeneity was quantified. Results Eleven pre-existing comorbidities from 25 studies were included in the meta-analysis (n = 65, 484 patients with COVID-19; mean age; 61 years; 57% male). Overall, the between-study heterogeneity was medium, and studies had low publication bias and high quality. Cardiovascular disease (risk ratio (RR) 2.25, 95% CI = 1.60–3.17, number of studies (n) = 14), hypertension (1.82 [1.43 to 2.32], n = 13), diabetes (1.48 [1.02 to 2.15], n = 16), congestive heart failure (2.03 [1.28 to 3.21], n = 3), chronic kidney disease (3.25 [1.13 to 9.28)], n = 9) and cancer (1.47 [1.01 to 2.14), n = 10) were associated with a significantly greater risk of mortality from COVID-19. Conclusions Patients with COVID-19 with cardiovascular disease, hypertension, diabetes, congestive heart failure, chronic kidney disease and cancer have a greater risk of mortality compared to patients with COVID-19 without these comorbidities. Tailored infection prevention and treatment strategies targeting this high-risk population might improve survival.

/pdf/association-of-cardiovascular-disease-and-10-other-pre-tegec4yatq.pdf

Association of cardiovascular disease and 10 other pre-existing comorbidities with COVID-19 mortality: A systematic review and meta-analysis.

Classification tasks in machine learning involving more than two classes are known by the name of "multi-class classification". Performance indicators are very useful when the aim is to evaluate and compare different classification models or machine learning techniques. Many metrics come in handy to test the ability of a multi-class classifier. Those metrics turn out to be useful at different stage of the development process, e.g. comparing the performance of two different models or analysing the behaviour of the same model by tuning different parameters. In this white paper we review a list of the most promising multi-class metrics, we highlight their advantages and disadvantages and show their possible usages during the development of a classification model.

Metrics for Multi-Class Classification: an Overview

Logistic regression analysis is a statistical technique to evaluate the relationship between various predictor variables (either categorical or continuous) and an outcome which is binary (dichotomous). In this article, we discuss logistic regression analysis and the limitations of this technique.

Common pitfalls in statistical analysis: Logistic regression.

Agreement between measurements refers to the degree of concordance between two (or more) sets of measurements. Statistical methods to test agreement are used to assess inter-rater variability or to decide whether one technique for measuring a variable can substitute another. In this article, we look at statistical measures of agreement for different types of data and discuss the differences between these and those for assessing correlation.

Common pitfalls in statistical analysis: Measures of agreement.

During the conduct of clinical trials, it is not uncommon to have protocol violations or inability to assess outcomes. This article in our series on common pitfalls in statistical analysis explains the complexities of analyzing results from such trials and highlights the importance of "intention-to-treat" analysis.

Common pitfalls in statistical analysis: Intention-to-treat versus per-protocol analysis.

In clinical research, study results, which are statistically significant are often interpreted as being clinically important. While statistical significance indicates the reliability of the study results, clinical significance reflects its impact on clinical practice. The third article in this series exploring pitfalls in statistical analysis clarifies the importance of differentiating between statistical significance and clinical significance.

Common pitfalls in statistical analysis: Clinical versus statistical significance

In biomedical research, we are often interested in quantifying the relationship between an exposure and an outcome. "Odds" and "Risk" are the most common terms which are used as measures of association between variables. In this article, which is the fourth in the series of common pitfalls in statistical analysis, we explain the meaning of risk and odds and the difference between the two.

Priya Ranganathan

Papers

Common pitfalls in statistical analysis: Logistic regression.

Common pitfalls in statistical analysis: Measures of agreement.

Common pitfalls in statistical analysis: Intention-to-treat versus per-protocol analysis.

Common pitfalls in statistical analysis: Clinical versus statistical significance

Common pitfalls in statistical analysis: Odds versus risk.