NEBULA is a fast negative binomial mixed model for differential or co-expression analysis of large-scale multi-subject single-cell data

The increasing availability of single-cell data revolutionizes the understanding of biological mechanisms at cellular resolution. For differential expression analysis in multi-subject single-cell data, negative binomial mixed models account for both subject-level and cell-level overdispersions, but are computationally demanding. Here, we propose an efficient NEgative Binomial mixed model Using a Large-sample Approximation (NEBULA). The speed gain is achieved by analytically solving high-dimensional integrals instead of using the Laplace approximation. We demonstrate that NEBULA is orders of magnitude faster than existing tools and controls false-positive errors in marker gene identification and co-expression analysis. Using NEBULA in Alzheimer’s disease cohort data sets, we found that the cell-level expression of APOE correlated with that of other genetic risk factors (including CLU, CST3, TREM2, C1q, and ITM2B) in a cell-type-specific pattern and an isoform-dependent manner in microglia. NEBULA opens up a new avenue for the broad application of mixed models to large-scale multi-subject single-cell data.

Linking the oral microbiome and salivary cytokine abundance to circadian oscillations

Saliva has immense potential as a diagnostic fluid for identification and monitoring of several systemic diseases. Composition of the microbiome and inflammation has been associated and reflective of oral and overall health. In addition, the relative ease of collection of saliva further strengthens large-scale diagnostic purposes. However, the future clinical utility of saliva cannot be fully determined without a detailed examination of daily fluctuations that may occur within the oral microbiome and inflammation due to circadian rhythm. In this study, we explored the association between the salivary microbiome and the concentration of IL-1β, IL-6 and IL-8 in the saliva of 12 healthy adults over a period of 24 h by studying the 16S rRNA gene followed by negative binomial mixed model regression analysis. To determine the periodicity and oscillation patterns of both the oral microbiome and inflammation (represented by the cytokine levels), two of the twelve subjects were studied for three consecutive days. Our results indicate that the Operational Taxonomic Units (OTUs) belonging to Prevotella, SR1 and Ruminococcaceae are significantly associated to IL-1β while Prevotella and Granulicatella were associated with IL-8. Our findings have also revealed a periodicity of both the oral microbiome (OTUs) and inflammation (cytokine levels) with identifiable patterns between IL-1β and Prevotella, and IL-6 with Prevotella, Neisseria and Porphyromonas. We believe that this study represents the first measure and demonstration of simultaneous periodic fluctuations of cytokine levels and specific populations of the oral microbiome.

Effect of Ivermectin Treatment on the Frequency of Seizures in Persons with Epilepsy Infected with Onchocerca volvulus

A clinical trial performed in the Democratic Republic of Congo (DRC), among persons with epilepsy (PWE) infected with Onchocerca volvulus treated with anti-seizure medication suggested that ivermectin reduces the seizure frequency. We assessed the effect of ivermectin treatment on seizure frequency in PWE with and without anti-seizure medication in three onchocerciasis endemic areas (Maridi, South Sudan; Aketi, DRC; and Mahenge, Tanzania). Pre- and 3–5 months post-ivermectin microfilariae densities in skin snips and seizure frequency were assessed. After ivermectin, the median (IQR) percentage reduction in seizure frequency in the study sites ranged from 73.4% (26.0–90.0) to 100% (50.0–100.0). A negative binomial mixed model showed that ivermectin significantly reduced the seizure frequency, with a larger decrease in PWE with a high baseline seizure frequency. Mediation analysis showed that ivermectin reduced the seizure frequencies indirectly through reduction in microfilariae densities but also that ivermectin may have a direct anti-seizure effect. However, given the short half-life of ivermectin and the fact that ivermectin does not penetrate the healthy brain, such a direct anti-seizure effect is unlikely. A randomized controlled trial assessing the ivermectin effect in people infected with O. volvulus who are also PWE on a stable anti-seizure regimen may be needed to clarify the causal relationship between ivermectin and seizure frequency.

NEBULA: a fast negative binomial mixed model for differential expression and co-expression analyses of large-scale multi-subject single-cell data

The growing availability of large-scale single-cell data revolutionizes our understanding of biological mechanisms at a finer resolution. In differential expression and co-expression analyses of multi-subject single-cell data, it is important to take into account both subject-level and cell-level overdispersions through negative binomial mixed models (NBMMs). However, the application of NBMMs to large-scale single-cell data is computationally demanding. In this work, we propose an efficient NEgative Binomial mixed model Using a Large-sample Approximation (NEBULA)), which analytically solves the high-dimensional integral in the marginal likelihood instead of using the Laplace approximation. Our benchmarks show that NEBULA dramatically reduces the running time by orders of magnitude compared to existing tools. We showed that NEBULA controlled false positives in identifying marker genes, while a simple negative binomial model produced spurious associations. Leveraging NEBULA, we decomposed between-subject and within-subject overdispersions of an snRNA-seq data set in the frontal cortex comprising ∼80,000 cells from a cohort of 48 individuals for Alzheimer’s diseases (AD). We observed that subpopulations and known subject-level covariates contributed substantially to the overdispersions. We carried out cell-type-specific transcriptome-wide within-subject co-expression analysis of APOE. The results revealed that APOE was most co-expressed with multiple AD-related genes, including CLU and CST3 in astrocytes, TREM2 and C1q genes in microglia, and ITM2B, an inhibitor of the amyloid-beta peptide aggregation, in both cell types. We found that the co-expression patterns were different in APOE2+ and APOE4+ cells in microglia, which suggest an isoform-dependent regulatory role in the immune system through the complement system in microglia. NEBULA opens up a new avenue for the broad application of NBMMs in the analysis of large-scale multi-subject single-cell data.

Zero-inflated negative binomial mixed model: an application to two microbial organisms important in oesophagitis

SUMMARYAltered microbial communities are thought to play an important role in eosinophilic oesophagitis, an allergic inflammatory condition of the oesophagus. Identification of the majority of organisms present in human-associated microbial communities is feasible with the advent of high throughput sequencing technology. However, these data consist of non-negative, highly skewed sequence counts with a large proportion of zeros. In addition, hierarchical study designs are often performed with repeated measurements or multiple samples collected from the same subject, thus requiring approaches to account for within-subject variation, yet only a small number of microbiota studies have applied hierarchical regression models. In this paper, we describe and illustrate the use of a hierarchical regression-based approach to evaluate multiple factors for a small number of organisms individually. More specifically, the zero-inflated negative binomial mixed model with random effects in both the count and zero-inflated parts is applied to evaluate associations with disease state while adjusting for potential confounders for two organisms of interest from a study of human microbiota sequence data in oesophagitis.

Application of zero-inflated negative binomial mixed model to human microbiota sequence data

Identification of the majority of organisms present in human-associated microbial communities is feasible with the advent of high throughput sequencing technology. However, these data consist of non-negative, highly skewed sequence counts with a large proportion of zeros. Zero-inflated models are useful for analyzing such data. Moreover, the non-zero observations may be over-dispersed in relation to the Poisson distribution, biasing parameter estimates and underestimating standard errors. In such a circumstance, a zero-inflated negative binomial (ZINB) model better accounts for these characteristics compared to a zero-inflated Poisson (ZIP). In addition, complex study designs are possible with repeated measurements or multiple samples collected from the same subject, thus random effects are introduced to account for the within subject variation. A zero-inflated negative binomial mixed model contains components to model the probability of excess zero values and the negative binomial parameters, allowing for repeated measures using independent random effects between these two components. The objective of this study is to examine the application of a zero-inflated negative binomial mixed model to human microbiota sequence data.

Application of zero-inflated negative binomial mixed model to human microbiota sequence data

Identification of the majority of organisms present in human-associated microbial communities is feasible with the advent of high throughput sequencing technology. However, these data consist of non-negative, highly skewed sequence counts with a large proportion of zeros. Zero-inflated models are useful for analyzing such data. Moreover, the non-zero observations may be over-dispersed in relation to the Poisson distribution, biasing parameter estimates and underestimating standard errors. In such a circumstance, a zero-inflated negative binomial (ZINB) model better accounts for these characteristics compared to a zero-inflated Poisson (ZIP). In addition, complex study designs are possible with repeated measurements or multiple samples collected from the same subject, thus random effects are introduced to account for the within subject variation. A zero-inflated negative binomial mixed model contains components to model the probability of excess zero values and the negative binomial parameters, allowing for repeated measures using independent random effects between these two components. The objective of this study is to examine the application of a zero-inflated negative binomial mixed model to human microbiota sequence data.