Trajectory inference across multiple conditions with condiments.

In single-cell RNA sequencing (scRNA-Seq), gene expression is assessed individually for each cell, allowing the investigation of developmental processes, such as embryogenesis and cellular differentiation and regeneration, at unprecedented resolution. In such dynamic biological systems, cellular states form a continuum, e.g., for the differentiation of stem cells into mature cell types. This process is often represented via a trajectory in a reduced-dimensional representation of the scRNA-Seq dataset. While many methods have been suggested for trajectory inference, it is often unclear how to handle multiple biological groups or conditions, e.g., inferring and comparing the differentiation trajectories of wild-type and knock-out stem cell populations. In this manuscript, we present condiments, a method for the inference and downstream interpretation of cell trajectories across multiple conditions. Our framework allows the interpretation of differences between conditions at the trajectory, cell population, and gene expression levels. We start by integrating datasets from multiple conditions into a single trajectory. By comparing the cell's conditions along the trajectory's path, we can detect large-scale changes, indicative of differential progression or fate selection. We also demonstrate how to detect subtler changes by finding genes that exhibit different behaviors between these conditions along a differentiation path.

A Latent Activated Olfactory Stem Cell State Revealed by Single Cell Transcriptomic and Epigenomic Profiling.

The olfactory epithelium is one of the few regions of the nervous system that sustains neurogenesis throughout life. Its experimental accessibility makes it especially tractable for studying molecular mechanisms that drive neural regeneration after injury-induced cell death. In this study, we used single cell sequencing to identify major regulatory players in determining olfactory epithelial stem cell fate after acute injury. We combined gene expression and accessible chromatin profiles of individual lineage traced olfactory stem cells to predict transcription factor activity specific to different lineages and stages of recovery. We further identified a discrete stem cell state that appears poised for activation, characterized by accessible chromatin around wound response and lineage specific genes prior to their later expression in response to injury. Together these results provide evidence that a subset of quiescent olfactory epithelial stem cells are epigenetically primed to support injury-induced regeneration.

HSA Adductomics Reveals Sex Differences in NHL Incidence and Possible Involvement of Microbial Translocation.

BACKGROUND: The higher incidence of non-Hodgkin lymphoma (NHL) in males is not well understood. Although reactive oxygen species (ROS) have been implicated as causes of NHL, they cannot be measured directly in archived blood. METHODS: We performed untargeted adductomics of stable ROS adducts in human serum albumin (HSA) from 67 incident NHL cases and 82 matched controls from the European Prospective Investigation into Cancer and Nutrition-Italy cohort. Regression and classification methods were employed to select features associated with NHL in all subjects and in males and females separately. RESULTS: Sixty seven HSA-adduct features were quantified by liquid chromatography-high-resolution mass spectrometry at Cys34 (n = 55) and Lys525 (n = 12). Three features were selected for association with NHL in all subjects, while seven were selected for males and five for females with minimal overlap. Two selected features were more abundant in cases and seven in controls, suggesting that altered homeostasis of ROS may affect NHL incidence. Heat maps revealed differential clustering of features between sexes, suggesting differences in operative pathways. CONCLUSIONS: Adduct clusters dominated by Cys34 oxidation products and disulfides further implicate ROS and redox biology in the etiology of NHL. Sex differences in dietary and alcohol consumption also help to explain the limited overlap of feature selection between sexes. Intriguingly, a disulfide of methanethiol from enteric microbial metabolism was more abundant in male cases, thereby implicating microbial translocation as a potential contributor to NHL in males. IMPACT: Only two of the ROS adducts associated with NHL overlapped between sexes and one adduct implicates microbial translocation as a risk factor.

Cross-Validated Loss-Based Covariance Matrix Estimator Selection in High Dimensions.

The covariance matrix plays a fundamental role in many modern exploratory and inferential statistical procedures, including dimensionality reduction, hypothesis testing, and regression. In low-dimensional regimes, where the number of observations far exceeds the number of variables, the optimality of the sample covariance matrix as an estimator of this parameter is well-established. High-dimensional regimes do not admit such a convenience. Thus, a variety of estimators have been derived to overcome the shortcomings of the canonical estimator in such settings. Yet, selecting an optimal estimator from among the plethora available remains an open challenge. Using the framework of cross-validated loss-based estimation, we develop the theoretical underpinnings of just such an estimator selection procedure. We propose a general class of loss functions for covariance matrix estimation and establish accompanying finite-sample risk bounds and conditions for the asymptotic optimality of the cross-validation selector. In numerical experiments, we demonstrate the optimality of our proposed selector in moderate sample sizes and across diverse data-generating processes. The practical benefits of our procedure are highlighted in a dimension reduction application to single-cell transcriptome sequencing data.

One-Carbon (Folate) Metabolism Pathway at Birth and Risk of Childhood Acute Lymphoblastic Leukemia: A Biomarker Study in Newborns.

Leukemia is the most common cancer in children in industrialized countries, and its initiation often occurs prenatally. Folic acid is a key vitamin in the production and modification of DNA, and prenatal folic acid intake is known to reduce the risk of childhood leukemia. We characterized the one-carbon (folate) metabolism nutrients that may influence risk of childhood acute lymphoblastic leukemia (ALL) among 122 cases diagnosed at age 0-14 years during 1988-2011 and 122 controls matched on sex, age, and race/ethnicity. Using hydrophilic interaction chromatography (HILIC) applied to neonatal dried blood spots, we evaluated 11 folate pathway metabolites, overall and by sex, race/ethnicity, and age at diagnosis. To conduct the prediction analyses, the 244 samples were separated into learning (75%) and test (25%) sets, maintaining the matched pairings. The learning set was used to train classification methods which were evaluated on the test set. High classification error rates indicate that the folate pathway metabolites measured have little predictive capacity for pediatric ALL. In conclusion, the one-carbon metabolism nutrients measured at birth were unable to predict subsequent leukemia in children. These negative findings are reflective of the last weeks of pregnancy and our study does not address the impact of these nutrients at the time of conception or during the first trimester of pregnancy that are critical for the embryo's DNA methylation programming.

A flexible approach for predictive biomarker discovery.

An endeavor central to precision medicine is predictive biomarker discovery; they define patient subpopulations which stand to benefit most, or least, from a given treatment. The identification of these biomarkers is often the byproduct of the related but fundamentally different task of treatment rule estimation. Using treatment rule estimation methods to identify predictive biomarkers in clinical trials where the number of covariates exceeds the number of participants often results in high false discovery rates. The higher than expected number of false positives translates to wasted resources when conducting follow-up experiments for drug target identification and diagnostic assay development. Patient outcomes are in turn negatively affected. We propose a variable importance parameter for directly assessing the importance of potentially predictive biomarkers and develop a flexible nonparametric inference procedure for this estimand. We prove that our estimator is double robust and asymptotically linear under loose conditions in the data-generating process, permitting valid inference about the importance metric. The statistical guarantees of the method are verified in a thorough simulation study representative of randomized control trials with moderate and high-dimensional covariate vectors. Our procedure is then used to discover predictive biomarkers from among the tumor gene expression data of metastatic renal cell carcinoma patients enrolled in recently completed clinical trials. We find that our approach more readily discerns predictive from nonpredictive biomarkers than procedures whose primary purpose is treatment rule estimation. An open-source software implementation of the methodology, the uniCATE R package, is briefly introduced.

Normalization benchmark of ATAC-seq datasets shows the importance of accounting for GC-content effects.

The assay for transposase-accessible chromatin using sequencing (ATAC-seq) allows the study of epigenetic regulation of gene expression by assessing chromatin configuration for an entire genome. Despite its popularity, there have been limited studies investigating the analytical challenges related to ATAC-seq data, with most studies leveraging tools developed for bulk transcriptome sequencing. Here, we show that GC-content effects are omnipresent in ATAC-seq datasets. Since the GC-content effects are sample specific, they can bias downstream analyses such as clustering and differential accessibility analysis. We introduce a normalization method based on smooth-quantile normalization within GC-content bins and evaluate it together with 11 different normalization procedures on 8 public ATAC-seq datasets. Accounting for GC-content effects in the normalization is crucial for common downstream ATAC-seq data analyses, improving accuracy and interpretability. Through case studies, we show that exploratory data analysis is essential to guide the choice of an appropriate normalization method for a given dataset.

CALDERA: finding all significant de Bruijn subgraphs for bacterial GWAS.

MOTIVATION: Genome-wide association studies (GWAS), aiming to find genetic variants associated with a trait, have widely been used on bacteria to identify genetic determinants of drug resistance or hypervirulence. Recent bacterial GWAS methods usually rely on k-mers, whose presence in a genome can denote variants ranging from single-nucleotide polymorphisms to mobile genetic elements. This approach does not require a reference genome, making it easier to account for accessory genes. However, a same gene can exist in slightly different versions across different strains, leading to diluted effects. RESULTS: Here, we overcome this issue by testing covariates built from closed connected subgraphs (CCSs) of the de Bruijn graph defined over genomic k-mers. These covariates capture polymorphic genes as a single entity, improving k-mer-based GWAS both in terms of power and interpretability. However, a method naively testing all possible subgraphs would be powerless due to multiple testing corrections, and the mere exploration of these subgraphs would quickly become computationally intractable. The concept of testable hypothesis has successfully been used to address both problems in similar contexts. We leverage this concept to test all CCSs by proposing a novel enumeration scheme for these objects which fully exploits the pruning opportunity offered by testability, resulting in drastic improvements in computational efficiency. Our method integrates with existing visual tools to facilitate interpretation. AVAILABILITY AND IMPLEMENTATION: We provide an implementation of our method, as well as code to reproduce all results at https://github.com/HectorRDB/Caldera_ISMB. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Extending the HSA-Cys34-Adductomics Pipeline to Modifications at Lys525.

We previously developed an adductomics pipeline that employed nanoflow liquid chromatography and high-resolution tandem mass spectrometry (nLC-HR-MS/MS) plus informatics to perform an untargeted detection of modifications to Cys34 in the tryptic T3 peptide of human serum albumin (HSA) (21ALVLIAFAQYLQQC34PFEDHVK41). In order to detect these peptide modifications without targeting specific masses, the pipeline interrogates MS2 ions that are signatures of the T3 peptide. The pipeline had been pilot-tested with archived plasma from healthy human subjects, and several of the 43 Cys34 adducts were highly associated with the smoking status. In the current investigation, we adapted the pipeline to include modifications to the ε-amino group of Lys525─a major glycation site in HSA─and thereby extend the coverage to products of Schiff bases that cannot be produced at Cys34. Because trypsin is generally unable to digest proteins at modified lysines, our pipeline detects miscleaved tryptic peptides with the sequence 525KQTALVELVK534. Adducts of both Lys525 and Cys34 are measured in a single nLC-HR-MS/MS run by increasing the mass range of precursor ions in MS1 scans and including both triply and doubly charged precursor ions for collision-induced dissociation fragmentation. For proof of principle, we applied the Cys34/Lys525 pipeline to archived plasma specimens from a subset of the same volunteer subjects used in the original investigation. Twelve modified Lys525 peptides were detected, including products of glycation (fructosyl-lysine plus advanced-glycated-end products), acetylation, and elimination of ammonia and water. Surprisingly, the carbamylated and glycated adducts were present at significantly lower levels in smoking subjects. By including a larger class of in vivo nucleophilic substitution reactions, the Cys34/Lys525 adductomics pipeline expands exposomic investigations of unknown human exposure to reactive electrophiles derived from both exogenous and endogenous sources.

Klf5 establishes bi-potential cell fate by dual regulation of ICM and TE specification genes.

Early blastomeres of mouse preimplantation embryos exhibit bi-potential cell fate, capable of generating both embryonic and extra-embryonic lineages in blastocysts. Here we identify three major two-cell-stage (2C)-specific endogenous retroviruses (ERVs) as the molecular hallmark of this bi-potential plasticity. Using the long terminal repeats (LTRs) of all three 2C-specific ERVs, we identify Krüppel-like factor 5 (Klf5) as their major upstream regulator. Klf5 is essential for bi-potential cell fate; a single Klf5-overexpressing embryonic stem cell (ESC) generates terminally differentiated embryonic and extra-embryonic lineages in chimeric embryos, and Klf5 directly induces inner cell mass (ICM) and trophectoderm (TE) specification genes. Intriguingly, Klf5 and Klf4 act redundantly during ICM specification, whereas Klf5 deficiency alone impairs TE specification. Klf5 is regulated by multiple 2C-specific transcription factors, particularly Dux, and the Dux/Klf5 axis is evolutionarily conserved. The 2C-specific transcription program converges on Klf5 to establish bi-potential cell fate, enabling a cell state with dual activation of ICM and TE genes.

A transcriptomic and epigenomic cell atlas of the mouse primary motor cortex.

Single-cell transcriptomics can provide quantitative molecular signatures for large, unbiased samples of the diverse cell types in the brain1-3. With the proliferation of multi-omics datasets, a major challenge is to validate and integrate results into a biological understanding of cell-type organization. Here we generated transcriptomes and epigenomes from more than 500,000 individual cells in the mouse primary motor cortex, a structure that has an evolutionarily conserved role in locomotion. We developed computational and statistical methods to integrate multimodal data and quantitatively validate cell-type reproducibility. The resulting reference atlas-containing over 56 neuronal cell types that are highly replicable across analysis methods, sequencing technologies and modalities-is a comprehensive molecular and genomic account of the diverse neuronal and non-neuronal cell types in the mouse primary motor cortex. The atlas includes a population of excitatory neurons that resemble pyramidal cells in layer 4 in other cortical regions4. We further discovered thousands of concordant marker genes and gene regulatory elements for these cell types. Our results highlight the complex molecular regulation of cell types in the brain and will directly enable the design of reagents to target specific cell types in the mouse primary motor cortex for functional analysis.

Untargeted metabolomics of newborn dried blood spots reveals sex-specific associations with pediatric acute myeloid leukemia.

The etiology of pediatric acute myeloid leukemia (AML) is largely unknown, but evidence for mutations in utero and long latency periods suggests that environmental factors play a role. Therefore, we used untargeted metabolomics of archived newborn dried blood spots (DBS) to investigate neonatal exposures as potential causal risk factors for AML. Untargeted metabolomics profiling was performed on DBS punches from 48 pediatric patients with AML and 46 healthy controls as part of the California Childhood Leukemia Study (CCLS). Because sex disparities are suggested by differences in AML incidence rates, metabolite features associated with AML were identified in analyses stratified by sex. There was no overlap between the 16 predictors of AML in females and 15 predictors in males, suggesting that neonatal metabolomic profiles of pediatric AML risk are sex-specific. In females, four predictors of AML were putatively annotated as ceramides, a class of metabolites that has been linked with cancer cell proliferation. In females, two metabolite predictors of AML were strongly correlated with breastfeeding duration, indicating a possible biological link between this putative protective risk factor and childhood leukemia. In males, a heterogeneous metabolite profile of AML predictors was observed. Replication with larger participant numbers is required to validate findings.

Non-neuronal expression of SARS-CoV-2 entry genes in the olfactory system suggests mechanisms underlying COVID-19-associated anosmia.

Altered olfactory function is a common symptom of COVID-19, but its etiology is unknown. A key question is whether SARS-CoV-2 (CoV-2) - the causal agent in COVID-19 - affects olfaction directly, by infecting olfactory sensory neurons or their targets in the olfactory bulb, or indirectly, through perturbation of supporting cells. Here we identify cell types in the olfactory epithelium and olfactory bulb that express SARS-CoV-2 cell entry molecules. Bulk sequencing demonstrated that mouse, non-human primate and human olfactory mucosa expresses two key genes involved in CoV-2 entry, ACE2 and TMPRSS2. However, single cell sequencing revealed that ACE2 is expressed in support cells, stem cells, and perivascular cells, rather than in neurons. Immunostaining confirmed these results and revealed pervasive expression of ACE2 protein in dorsally-located olfactory epithelial sustentacular cells and olfactory bulb pericytes in the mouse. These findings suggest that CoV-2 infection of non-neuronal cell types leads to anosmia and related disturbances in odor perception in COVID-19 patients.

Exploring high-dimensional biological data with sparse contrastive principal component analysis.

MOTIVATION: Statistical analyses of high-throughput sequencing data have re-shaped the biological sciences. In spite of myriad advances, recovering interpretable biological signal from data corrupted by technical noise remains a prevalent open problem. Several classes of procedures, among them classical dimensionality reduction techniques and others incorporating subject-matter knowledge, have provided effective advances. However, no procedure currently satisfies the dual objectives of recovering stable and relevant features simultaneously. RESULTS: Inspired by recent proposals for making use of control data in the removal of unwanted variation, we propose a variant of principal component analysis (PCA), sparse contrastive PCA that extracts sparse, stable, interpretable and relevant biological signal. The new methodology is compared to competing dimensionality reduction approaches through a simulation study and via analyses of several publicly available protein expression, microarray gene expression and single-cell transcriptome sequencing datasets. AVAILABILITY AND IMPLEMENTATION: A free and open-source software implementation of the methodology, the scPCA R package, is made available via the Bioconductor Project. Code for all analyses presented in this article is also available via GitHub. CONTACT: philippe_boileau@berkeley.edu. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Trajectory-based differential expression analysis for single-cell sequencing data.

Trajectory inference has radically enhanced single-cell RNA-seq research by enabling the study of dynamic changes in gene expression. Downstream of trajectory inference, it is vital to discover genes that are (i) associated with the lineages in the trajectory, or (ii) differentially expressed between lineages, to illuminate the underlying biological processes. Current data analysis procedures, however, either fail to exploit the continuous resolution provided by trajectory inference, or fail to pinpoint the exact types of differential expression. We introduce tradeSeq, a powerful generalized additive model framework based on the negative binomial distribution that allows flexible inference of both within-lineage and between-lineage differential expression. By incorporating observation-level weights, the model additionally allows to account for zero inflation. We evaluate the method on simulated datasets and on real datasets from droplet-based and full-length protocols, and show that it yields biological insights through a clear interpretation of the data.

Untargeted adductomics of newborn dried blood spots identifies modifications to human serum albumin associated with childhood leukemia.

The developing fetus is exposed to chemicals, which are metabolized to electrophiles that form adducts with nucleophilic Cys34 of human serum albumin (HSA). By measuring these adducts in neonatal blood spots (NBS), we obtain information regarding fetal exposures during the last month of gestation. To discover potential risk factors for childhood leukemia resulting from in utero exposures, we used untargeted adductomics to measure HSA-Cys34 adducts in 782 archived NBS, collected from incident cases of childhood acute lymphoblastic leukemia (ALL) or acute myeloid leukemia (AML) and matched population-based controls. Among a total of 28 Cys34 modifications that were measured, we found no differences in adduct abundances between childhood leukemia cases and controls overall. However, cases of T-cell ALL had higher abundances of adducts of reactive carbonyl species and a Cys34 disulfide of homocysteine was present at lower levels in AML cases. These results suggest that oxidative stress and lipid peroxidation may be etiologic factors of T-cell ALL, and alterations in one-carbon metabolism and epigenetic changes may be predictors of AML. Future replication of the results with larger sample sizes is necessary.

Cys34 Adductomics Links Colorectal Cancer with the Gut Microbiota and Redox Biology.

Chronic inflammation is an established risk factor for colorectal cancer. To study reactive products of gut inflammation and redox signaling on colorectal cancer development, we used untargeted adductomics to detect adduct features in prediagnostic serum from the EPIC Italy cohort. We focused on modifications to Cys34 in human serum albumin, which is responsible for scavenging small reactive electrophiles that might initiate cancers. Employing a combination of statistical methods, we selected seven Cys34 adducts associated with colorectal cancer, as well as body mass index (BMI; a well-known risk factor). Five adducts were more abundant in colorectal cancer cases than controls and clustered with each other, suggesting a common pathway. Because two of these adducts were Cys34 modifications by methanethiol, a microbial-human cometabolite, and crotonaldehyde, a product of lipid peroxidation, these findings further implicate infiltration of gut microbes into the intestinal mucosa and the corresponding inflammatory response as causes of colorectal cancer. The other two associated adducts were Cys34 disulfides of homocysteine that were less abundant in colorectal cancer cases than controls and may implicate homocysteine metabolism as another causal pathway. The selected adducts and BMI ranked higher as potentially causal factors than variables previously associated with colorectal cancer (smoking, alcohol consumption, physical activity, and total meat consumption). Regressions of case-control differences in adduct levels on days to diagnosis showed no statistical evidence that disease progression, rather than causal factors at recruitment, contributed to the observed differences. These findings support the hypothesis that infiltration of gut microbes into the intestinal mucosa and the resulting inflammation are causal factors for colorectal cancer. SIGNIFICANCE: Infiltration of gut microbes into the intestinal mucosa and the resulting inflammation are causal factors for colorectal cancer.

Filtering procedures for untargeted LC-MS metabolomics data.

BACKGROUND: Untargeted metabolomics datasets contain large proportions of uninformative features that can impede subsequent statistical analysis such as biomarker discovery and metabolic pathway analysis. Thus, there is a need for versatile and data-adaptive methods for filtering data prior to investigating the underlying biological phenomena. Here, we propose a data-adaptive pipeline for filtering metabolomics data that are generated by liquid chromatography-mass spectrometry (LC-MS) platforms. Our data-adaptive pipeline includes novel methods for filtering features based on blank samples, proportions of missing values, and estimated intra-class correlation coefficients. RESULTS: Using metabolomics datasets that were generated in our laboratory from samples of human blood, as well as two public LC-MS datasets, we compared our data-adaptive filtering method with traditional methods that rely on non-method specific thresholds. The data-adaptive approach outperformed traditional approaches in terms of removing noisy features and retaining high quality, biologically informative ones. The R code for running the data-adaptive filtering method is provided at https://github.com/courtneyschiffman/Metabolomics-Filtering . CONCLUSIONS: Our proposed data-adaptive filtering pipeline is intuitive and effectively removes uninformative features from untargeted metabolomics datasets. It is particularly relevant for interrogation of biological phenomena in data derived from complex matrices associated with biospecimens.

Performance Assessment and Selection of Normalization Procedures for Single-Cell RNA-Seq.

Systematic measurement biases make normalization an essential step in single-cell RNA sequencing (scRNA-seq) analysis. There may be multiple competing considerations behind the assessment of normalization performance, of which some may be study specific. We have developed "scone"- a flexible framework for assessing performance based on a comprehensive panel of data-driven metrics. Through graphical summaries and quantitative reports, scone summarizes trade-offs and ranks large numbers of normalization methods by panel performance. The method is implemented in the open-source Bioconductor R software package scone. We show that top-performing normalization methods lead to better agreement with independent validation data for a collection of scRNA-seq datasets. scone can be downloaded at http://bioconductor.org/packages/scone/.

Metabolomics of neonatal blood spots reveal distinct phenotypes of pediatric acute lymphoblastic leukemia and potential effects of early-life nutrition.

Early-life exposures are believed to influence the incidence of pediatric acute lymphoblastic leukemia (ALL). Archived neonatal blood spots (NBS), collected within the first days of life, offer a means to investigate small molecules that reflect early-life exposures. Using untargeted metabolomics, we compared abundances of small-molecule features in extracts of NBS punches from 332 children that later developed ALL and 324 healthy controls. Subjects were stratified by early (1-5 y) and late (6-14 y) diagnosis. Mutually-exclusive sets of metabolic features - representing putative lipids and fatty acids - were associated with ALL, including 9 and 19 metabolites in the early- and late-diagnosis groups, respectively. In the late-diagnosis group, a prominent cluster of features with apparent 18:2 fatty-acid chains suggested that newborn exposure to the essential nutrient, linoleic acid, increased ALL risk. Interestingly, abundances of these putative 18:2 lipids were greater in infants who were fed formula rather than breast milk (colostrum) and increased with the mother's pre-pregnancy body mass index. These results suggest possible etiologic roles of newborn nutrition in late-diagnosis ALL.