Batch Effects Research Articles

LMD: Cluster-Independent Multiscale Marker Identification in Single-cell RNA-seq Data.

Identifying accurate cell markers in single-cell RNA-seq data is crucial for understanding cellular diversity and function. Localized Marker Detector (LMD) is a novel tool to identify "localized genes" - genes exclusively expressed in groups of highly similar cells - thereby characterizing cellular diversity in a multi-resolution and fine-grained manner. LMD constructs a cell-cell affinity graph, diffuses the gene expression value across the cell graph, and assigns a score to each gene based on its diffusion dynamics. LMD's candidate markers can be grouped into functional gene modules, which accurately reflect cell types, subtypes, and other sources of variation such as cell cycle status. We apply LMD to mouse bone marrow and hair follicle dermal condensate datasets, where LMD facilitates cross-sample comparisons, identifying shared and sample-specific gene signatures and novel cell populations without requiring batch effect correction or integration methods. Furthermore, we assessed the performance of LMD across nine single-cell RNA sequencing datasets, compared it with six other methods aimed at achieving similar objectives, and found that LMD outperforms the other methods evaluated.

Correction of Batch Effect in Gut Microbiota Profiling of ASD Cohorts from Different Geographical Origins.

To date, there have been numerous metataxonomic studies on gut microbiota (GM) profiling based on the analyses of data from public repositories. However, differences in study population and wet and dry pipelines have produced discordant results. Herein, we propose a biostatistical approach to remove these batch effects for the GM characterization in the case of autism spectrum disorders (ASDs). An original dataset of GM profiles from patients with ASD was ecologically characterized and compared with GM public digital profiles of age-matched neurotypical controls (NCs). Also, GM data from seven case-control studies on ASD were retrieved from the NCBI platform and exploited for analysis. Hence, on each dataset, conditional quantile regression (CQR) was performed to reduce the batch effects originating from both technical and geographical confounders affecting the GM-related data. This method was further applied to the whole dataset matrix, obtained by merging all datasets. The ASD GM markers were identified by the random forest (RF) model. We observed a different GM profile in patients with ASD compared with NC subjects. Moreover, a significant reduction of technical- and geographical-dependent batch effects in all datasets was achieved. We identified Bacteroides_H, Faecalibacterium, Gemmiger_A_73129, Blautia_A_141781, Bifidobacterium_388775, and Phocaeicola_A_858004 as robust GM bacterial biomarkers of ASD. Finally, our validation approach provided evidence of the validity of the QCR method, showing high values of accuracy, specificity, sensitivity, and AUC-ROC. Herein, we proposed an updated biostatistical approach to reduce the technical and geographical batch effects that may negatively affect the description of bacterial composition in microbiota studies.

DGSIST: Clustering spatial transcriptome data based on deep graph structure Infomax

Although spatial transcriptomics data provide valuable insights into gene expression profiles and the spatial structure of tissues, most studies rely solely on gene expression information, underutilizing the spatial data. To fully leverage the potential of spatial transcriptomics and graph neural networks, the DGSI (Deep Graph Structure Infomax) model is proposed. This innovative graph data processing model uses graph convolutional neural networks and employs an unsupervised learning approach. It maximizes the mutual information between graph-level and node-level representations, emphasizing flexible sampling and aggregation of nodes and their neighbors. This effectively captures and incorporates local information from nodes into the overall graph structure. Additionally, this paper developed the DGSIST framework, an unsupervised cell clustering method that integrates the DGSI model, SVD dimensionality reduction algorithm, and k-means++ clustering algorithm. This aims to identify cell types accurately. DGSIST fully uses spatial transcriptomics data and outperforms existing methods in accuracy. Demonstrations of DGSIST’s capability across various tissue types and technological platforms have shown its effectiveness in accurately identifying spatial domains in multiple tissue sections. Compared to other spatial clustering methods, DGSIST excels in cell clustering and effectively eliminates batch effects without needing batch correction. DGSIST excels in spatial clustering analysis, spatial variation identification, and differential gene expression detection and directly applies to graph analysis tasks, such as node classification, link prediction, or graph clustering. Anticipation lies in the contribution of the DGSIST framework to a deeper understanding of the spatial organizational structures of diseases such as cancer.

MitoSort: Robust Demultiplexing of Pooled Single-cell Genomics Data Using Endogenous Mitochondrial Variants.

Multiplexing across donors has emerged as a popular strategy to increase throughput, reduce costs, overcome technical batch effects, and improve doublet detection in single-cell genomic studies. To eliminate additional experimental steps, endogenous nuclear genome variants are used for demultiplexing pooled single-cell RNA sequencing (scRNA-seq) data by several computational tools. However, these tools have limitations when applied to single-cell sequencing methods that do not cover nuclear genomic regions well, such as single-cell assay for transposase-accessible chromatin with sequencing (scATAC-seq). Here, we demonstrate that mitochondrial germline variants are an alternative, robust, and computationally efficient endogenous barcode for sample demultiplexing. We propose MitoSort, a tool that uses mitochondrial germline variants to assign cells to their donor of origin and identify cross-genotype doublets in single-cell genomics datasets. We evaluate its performance by using in silico pooled mitochondrial scATAC-seq (mtscATAC-seq) libraries and experimentally multiplexed data with cell hashtags. MitoSort achieves high accuracy and efficiency in genotype clustering and doublet detection for mtscATAC-seq data, addressing the limitations of current computational techniques tailored for scRNA-seq data. Moreover, MitoSort exhibits versatility and can be applied to various single-cell sequencing approaches beyond mtscATAC-seq, provided the mitochondrial variants are reliably detected. Furthermore, we demonstrate the application of MitoSort in a case study where B cells from eight donors were pooled and assayed by single-cell multi-omics sequencing. Altogether, our results demonstrate the accuracy and efficiency of MitoSort, which enables reliable sample demultiplexing in various single-cell genomic applications. MitoSort is available at https://github.com/tangzhj/MitoSort.

Systematic benchmark of single-cell hashtag demultiplexing approaches reveals robust performance of a clustering-based method.

Single-cell technology opened up a new avenue to delineate cellular status at a single-cell resolution and has become an essential tool for studying human diseases. Multiplexing allows cost-effective experiments by combining multiple samples and effectively mitigates batch effects. It starts by giving each sample a unique tag and then pooling them together for library preparation and sequencing. After sequencing, sample demultiplexing is performed based on tag detection, where cells belonging to one sample are expected to have a higher amount of the corresponding tag than cells from other samples. However, in reality, demultiplexing is not straightforward due to the noise and contamination from various sources. Successful demultiplexing depends on the efficient removal of such contamination. Here, we perform a systematic benchmark combining different normalization methods and demultiplexing approaches using real-world data and simulated datasets. We show that accounting for sequencing depth variability increases the separability between tagged and untagged cells, and the clustering-based approach outperforms existing tools. The clustering-based workflow is available as an R package from https://github.com/hwlim/hashDemux.

6527 Genetic variants may influence the glycemic response to sodium-glucose cotransporter-2 inhibitors in the Mass General Brigham Biobank

Abstract Disclosure: J.H. Li: None. A. Barry: None. A. Huerta-Chagoya: None. L. Szczerbinski: None. J.M. Mercader: None. J.C. Florez: Consulting Fee; Self; AstraZeneca. Grant Recipient; Self; Novo Nordisk. Speaker; Self; Novo Nordisk, Merck. A. Leong: None. Background: Sodium-glucose cotransporter-2 inhibitors (SGLT2i) are increasingly prescribed for the treatment of type 2 diabetes, but there is interindividual variability in glycemic response. We investigated whether two genetic variants, rs33954001, a missense variant (p.His615Gln) in SLC5A1 encoding SGLT1, and rs8050500 near SLC5A2 encoding SGLT2, previously associated with glycemic traits in genome-wide association studies, had any influence on short-term glycated hemoglobin (HbA1c) lowering in patients on SGLT2i treatment. Methods: We identified 1,961 ancestrally diverse, genotyped individuals with documented exposure to a SGLT2i in the Mass General Brigham Biobank, a hospital-based biobank linked with patient medical records. Of these, 736 individuals had a baseline HbA1c ≥ 6% within 6 months of the index appearance of an SGLT2i and a follow-up HbA1c measured 6-12 months after. Genotyping was performed on the Illumina Multi-Ethnic Genotyping Array and the Illumina Global Screening Array. Variants were imputed with the TOPMed reference panel. Using linear regression models, we evaluated the additive effect of the genetic variants on HbA1c reduction, defined as baseline HbA1c minus follow-up HbA1c, adjusted for age, sex, 10 genetic ancestry principal components to account for population stratification, the number of concomitant glucose-lowering drug classes, and three covariates accounting for potential genotyping batch effects. To determine whether genetic effects varied by pretreatment glycemic control, we tested for a genotype × baseline HbA1c interaction and performed stratified analysis by baseline HbA1c (≥ 8% vs. < 8%). Results: The mean age of participants was 63.7 years and 41% were female. The baseline HbA1c was 8.0 ± 1.9%. The mean HbA1c reduction was 0.6% over a mean follow-up time of 8.7 ± 1.6 months. None of the variants were associated with HbA1c reduction (p > 0.05) in the whole dataset. However, in stratified analyses, among those with baseline HbA1c ≥ 8% (n = 382), we observed that carriers of the G allele at rs33954001 had a 0.5% greater HbA1c reduction (p = 0.048) compared to those with CC genotype. Further adjustment for baseline renal function did not modify results. In the Genotype-Tissue Expression (GTEx) portal, the G allele was associated with higher SLC5A1 expression in nerve (p = 1 × 10-8), esophageal mucosa (p = 4 × 10-5) and stomach (p = 9 × 10-5), but not kidney. In the Translational human pancreatic Islet Genotype tissue-Expression Resource (TIGER), the G allele was associated with SLC5A1 expression in pancreatic islets (p = 2 × 10-6). Conclusion: Genetic variation in SLC5A1 may partly explain heterogeneity in SGLT2i treatment effects among people with sub-optimally controlled diabetes, possibly due to differences in SLC5A1 tissue-specific expression. These findings require validation in large hospital-based biobanks or dedicated clinical trials. Presentation: 6/2/2024

GEDI: An R Package for Integration of Transcriptomic Data from Multiple Platforms for Bioinformatics Applications.

Transcriptomic data is often expensive and difficult to generate in large cohorts relative to genomic data; therefore, it is often important to integrate multiple transcriptomic datasets from both microarray- and next generation sequencing (NGS)-based transcriptomic data across similar experiments or clinical trials to improve analytical power and discovery of novel transcripts and genes. However, transcriptomic data integration presents a few challenges including reannotation and batch effect removal. We developed the Gene Expression Data Integration (GEDI) R package to enable transcriptomic data integration by combining existing R packages. With just four functions, the GEDI R package makes constructing a transcriptomic data integration pipeline straightforward. Together, the functions overcome the complications in transcriptomic data integration by automatically reannotating the data and removing the batch effect. The removal of the batch effect is verified with principal component analysis and the data integration is verified using a logistic regression model with forward stepwise feature selection. To demonstrate the functionalities of the GEDI package, we integrated five bovine endometrial transcriptomic datasets from the NCBI Gene Expression Omnibus. These transcriptomic datasets were from multiple high-throughput platforms, namely, array-based Affymetrix and Agilent platforms, and NGS-based Illumina paired-end RNA-seq platform. Furthermore, we compared the GEDI package to existing tools and found that GEDI is the only tool that provides a full transcriptomic data integration pipeline including verification of both batch effect removal and data integration for downstream genomic and bioinformatics applications. © 2024 The Author(s). Current Protocols published by Wiley Periodicals LLC. Basic Protocol 1: ReadGE, a function to import gene expression datasets Basic Protocol 2: GEDI, a function to reannotate and merge gene expression datasets Basic Protocol 3: BatchCorrection, a function to remove batch effects from gene expression data Basic Protocol 4: VerifyGEDI, a function to confirm successful integration of gene expression data.

Statistical batch-aware embedded integration, dimension reduction, and alignment for spatial transcriptomics.

Spatial transcriptomics (ST) technologies provide richer insights into the molecular characteristics of cells by simultaneously measuring gene expression profiles and their relative locations. However, each slice can only contain limited biological variation, and since there are almost always non-negligible batch effects across different slices, integrating numerous slices to account for batch effects and locations is not straightforward. Performing multi-slice integration, dimensionality reduction, and other downstream analyses separately often results in suboptimal embeddings for technical artifacts and biological variations. Joint modeling integrating these steps can enhance our understanding of the complex interplay between technical artifacts and biological signals, leading to more accurate and insightful results. In this context, we propose a hierarchical hidden Markov random field model STADIA to reduce batch effects, extract common biological patterns across multiple ST slices, and simultaneously identify spatial domains. We demonstrate the effectiveness of STADIA using five datasets from different species (human and mouse), various organs (brain, skin, and liver), and diverse platforms (10x Visium, ST, and Slice-seqV2). STADIA can capture common tissue structures across multiple slices and preserve slice-specific biological signals. In addition, STADIA outperforms the other three competing methods (PRECAST, fastMNN, and Harmony) in terms of the balance between batch mixing and spatial domain identification, and it demonstrates the advantage of joint modeling when compared to STAGATE and GraphST. The source code implemented by R is available at https://github.com/zhanglabtools/STADIA and archived with version 1.01 on Zenodo https://zenodo.org/records/13637744.

Data-driven batch detection enhances single-cell omics data analysis

In single-cell omics studies, data are typically collected across multiple batches, resulting in batch effects: technical confounders that introduce noise and distort data distribution. Correcting these effects is challenging due to their unknown sources, nonlinear distortions, and the difficulty of accurately assigning data to batches that are optimal for integration methods.

BatchFLEX: feature-level equalization of X-batch.

Integrative analysis of heterogeneous expression data remains challenging due to variations in platform, RNA quality, sample processing, and other unknown technical effects. Selecting the approach for removing unwanted batch effects can be a time-consuming and tedious process, especially for more biologically focused investigators. Here, we present BatchFLEX, a Shiny app that can facilitate visualization and correction of batch effects using several established methods. BatchFLEX can visualize the variance contribution of a factor before and after correction. As an example, we have analyzed ImmGen microarray data and enhanced its expression signals that distinguishes each immune cell type. Moreover, our analysis revealed the impact of the batch correction in altering the gene expression rank and single-sample GSEA pathway scores in immune cell types, highlighting the importance of real-time assessment of the batch correction for optimal downstream analysis. Our tool is available through Github https://github.com/shawlab-moffitt/BATCH-FLEX-ShinyApp with an online example on Shiny.io https://shawlab-moffitt.shinyapps.io/batch_flex/.

GraphCVAE: Uncovering cell heterogeneity and therapeutic target discovery through residual and contrastive learning.

Advancements in Spatial Transcriptomics (ST) technologies in recent years have transformed the analysis of tissue structure and function within spatial contexts. However, accurately identifying spatial domains remains challenging due to data sparsity and noise. Traditional clustering methods often fail to capture spatial dependencies, while spatial clustering methods struggle with batch effects and data integration. We introduce GraphCVAE, a model designed to enhance spatial domain identification by integrating spatial and morphological information, correcting batch effects, and managing heterogeneous data. GraphCVAE employs a multi-layer Graph Convolutional Network (GCN) and a variational autoencoder to improve the representation and integration of spatial information. Through contrastive learning, the model captures subtle differences between cell types and states. Extensive testing on various ST datasets demonstrates GraphCVAE's robustness and biological contributions. In the dorsolateral prefrontal cortex (DLPFC) dataset, it accurately delineates cortical layer boundaries. In glioblastoma, GraphCVAE reveals critical therapeutic targets such as TF and NFIB. In colorectal cancer, it explores the role of the extracellular matrix in colorectal cancer. The model's performance metrics consistently surpass existing methods, validating its effectiveness. GraphCVAE's advanced visualization capabilities further highlight its precision in resolving spatial structures, making it a powerful tool for spatial transcriptomics analysis and offering new insights into disease studies.

Pairwise analysis of gene expression for oral squamous cell carcinoma via a large-scale transcriptome integration.

Among all cancers occurring in the head and neck region, oral squamous cell carcinoma (OSCC) is the most common oral malignant tumours characterized by its aggressiveness and metastasis. The development of transcriptomics technology has greatly facilitated the diagnosis of various cancers. However, identifying genetic biomarkers is limited by data from a single batch of OSCC samples, and integrating analysis across different platforms remains a great challenge. In this study, we integrated five OSCC transcriptome datasets using an innovative strategy capable of mitigating batch effect, and extracting information from different datasets based on changes in the relative expression of gene pairs. By leveraging a machine learning method, we developed a prediction model including 27 differential gene pairs (DGPs) to discriminate OSCC from control samples, achieving an area under the receiver operating characteristic curve (AUC) of 0.8987 for the training set. Moreover, the model demonstrated commendable performance in four external validation sets, with AUCs of 0.9926, 0.9688, 0.8052 and 0.8565, respectively. Subsequently, a prognostic model was constructed based on six key gene pairs through univariate and multivariate Cox regression analysis. The AUCs of the model at 1-year and 3-year overall survival time prediction were 0.717 and 0.779 in an independent dataset. Our result demonstrates the effectiveness of this new method of integrating data and identifying DGPs. Using DGPs can significantly improve the performance of both diagnostic and prognostic models.

Biomarker discovery in progressive supranuclear palsy from human cerebrospinal fluid

BackgroundProgressive supranuclear palsy (PSP) is a neurodegenerative disorder often misdiagnosed as Parkinson’s Disease (PD) due to shared symptoms. PSP is characterized by the accumulation of tau protein in specific brain regions, leading to loss of balance, gaze impairment, and dementia. Diagnosing PSP is challenging, and there is a significant demand for reliable biomarkers. Existing biomarkers, including tau protein and neurofilament light chain (NfL) levels in cerebrospinal fluid (CSF), show inconsistencies in distinguishing PSP from other neurodegenerative disorders. Therefore, the development of new biomarkers for PSP is imperative.MethodsWe conducted an extensive proteome analysis of CSF samples from 40 PSP patients, 40 PD patients, and 40 healthy controls (HC) using tandem mass tag-based quantification. Mass spectrometry analysis of 120 CSF samples was performed across 13 batches of 11-plex TMT experiments, with data normalization to reduce batch effects. Pathway, interactome, cell-type-specific enrichment, and bootstrap receiver operating characteristic analyses were performed to identify key candidate biomarkers.ResultsWe identified a total of 3,653 unique proteins. Our analysis revealed 190, 152, and 247 differentially expressed proteins in comparisons of PSP vs. HC, PSP vs. PD, and PSP vs. both PD and HC, respectively. Gene set enrichment and interactome analysis of the differentially expressed proteins in PSP CSF showed their involvement in cell adhesion, cholesterol metabolism, and glycan biosynthesis. Cell-type enrichment analysis indicated a predominance of neuronally-derived proteins among the differentially expressed proteins. The potential biomarker classification performance demonstrated that ATP6AP2 (reduced in PSP) had the highest AUC (0.922), followed by NEFM, EFEMP2, LAMP2, CHST12, FAT2, B4GALT1, LCAT, CBLN3, FSTL5, ATP6AP1, and GGH.ConclusionBiomarker candidate proteins ATP6AP2, NEFM, and CHI3L1 were identified as key differentiators of PSP from the other groups. This study represents the first large-scale use of mass spectrometry-based proteome analysis to identify cerebrospinal fluid (CSF) biomarkers specific to progressive supranuclear palsy (PSP) that can differentiate it from Parkinson’s disease (PD) and healthy controls. Our findings lay a crucial foundation for the development and validation of reliable biomarkers, which will enhance diagnostic accuracy and facilitate early detection of PSP.

Thinking points for effective batch correction on biomedical data.

Batch effects introduce significant variability into high-dimensional data, complicating accurate analysis and leading to potentially misleading conclusions if not adequately addressed. Despite technological and algorithmic advancements in biomedical research, effectively managing batch effects remains a complex challenge requiring comprehensive considerations. This paper underscores the necessity of a flexible and holistic approach for selecting batch effect correction algorithms (BECAs), advocating for proper BECA evaluations and consideration of artificial intelligence-based strategies. We also discuss key challenges in batch effect correction, including the importance of uncovering hidden batch factors and understanding the impact of design imbalance, missing values, and aggressive correction. Our aim is to provide researchers with a robust framework for effective batch effects management and enhancing the reliability of high-dimensional data analyses.

SELF-Former: multi-scale gene filtration transformer for single-cell spatial reconstruction.

The spatial reconstruction of single-cell RNA sequencing (scRNA-seq) data into spatial transcriptomics (ST) is a rapidly evolving field that addresses the significant challenge of aligning gene expression profiles to their spatial origins within tissues. This task is complicated by the inherent batch effects and the need for precise gene expression characterization to accurately reflect spatial information. To address these challenges, we developed SELF-Former, a transformer-based framework that utilizes multi-scale structures to learn gene representations, while designing spatial correlation constraints for the reconstruction of corresponding ST data. SELF-Former excels in recovering the spatial information of ST data and effectively mitigates batch effects between scRNA-seq and ST data. A novel aspect of SELF-Former is the introduction of a gene filtration module, which significantly enhances the spatial reconstruction task by selecting genes that are crucial for accurate spatial positioning and reconstruction. The superior performance and effectiveness of SELF-Former's modules have been validated across four benchmark datasets, establishing it as a robust and effective method for spatial reconstruction tasks. SELF-Former demonstrates its capability to extract meaningful gene expression information from scRNA-seq data and accurately map it to the spatial context of real ST data. Our method represents a significant advancement in the field, offering a reliable approach for spatial reconstruction.

Novel multi-omics deconfounding variational autoencoders can obtain meaningful disease subtyping.

Unsupervised learning, particularly clustering, plays a pivotal role in disease subtyping and patient stratification, especially with the abundance of large-scale multi-omics data. Deep learning models, such as variational autoencoders (VAEs), can enhance clustering algorithms by leveraging inter-individual heterogeneity. However, the impact of confounders-external factors unrelated to the condition, e.g. batch effect or age-on clustering is often overlooked, introducing bias and spurious biological conclusions. In this work, we introduce four novel VAE-based deconfounding frameworks tailored for clustering multi-omics data. These frameworks effectively mitigate confounding effects while preserving genuine biological patterns. The deconfounding strategies employed include (i) removal of latent features correlated with confounders, (ii) a conditional VAE, (iii) adversarial training, and (iv) adding a regularization term to the loss function. Using real-life multi-omics data from The Cancer Genome Atlas, we simulated various confounding effects (linear, nonlinear, categorical, mixed) and assessed model performance across 50 repetitions based on reconstruction error, clustering stability, and deconfounding efficacy. Our results demonstrate that our novel models, particularly the conditional multi-omics VAE (cXVAE), successfully handle simulated confounding effects and recover biologically driven clustering structures. cXVAE accurately identifies patient labels and unveils meaningful pathological associations among cancer types, validating deconfounded representations. Furthermore, our study suggests that some of the proposed strategies, such as adversarial training, prove insufficient in confounder removal. In summary, our study contributes by proposing innovative frameworks for simultaneous multi-omics data integration, dimensionality reduction, and deconfounding in clustering. Benchmarking on open-access data offers guidance to end-users, facilitating meaningful patient stratification for optimized precision medicine.

Structure-preserved integration of scRNA-seq data using heterogeneous graph neural network.

The integration of single-cell RNA sequencing (scRNA-seq) data from multiple experimental batches enables more comprehensive characterizations of cell states. Given that existing methods disregard the structural information between cells and genes, we proposed a structure-preserved scRNA-seq data integration approach using heterogeneous graph neural network (scHetG). By establishing a heterogeneous graph that represents the interactions between multiple batches of cells and genes, and combining a heterogeneous graph neural network with contrastive learning, scHetG concurrently obtained cell and gene embeddings with structural information. A comprehensive assessment covering different species, tissues and scales indicated that scHetG is an efficacious method for eliminating batch effects while preserving the structural information of cells and genes, including batch-specific cell types and cell-type specific gene co-expression patterns.

Single-cell mosaic integration and cell state transfer with auto-scaling self-attention mechanism.

The integration of data from multiple modalities generated by single-cell omics technologies is crucial for accurately identifying cell states. One challenge in comprehending multi-omics data resides in mosaic integration, in which different data modalities are profiled in different subsets of cells, as it requires simultaneous batch effect removal and modality alignment. Here, we develop Multi-omics Mosaic Auto-scaling Attention Variational Inference (mmAAVI), a scalable deep generative model for single-cell mosaic integration. Leveraging auto-scaling self-attention mechanisms, mmAAVI can map arbitrary combinations of omics to the common embedding space. If existing well-annotated cell states, the model can perform semisupervised learning to utilize existing these annotations. We validated the performance of mmAAVI and five other commonly used methods on four benchmark datasets, which vary in cell numbers, omics types, and missing patterns. mmAAVI consistently demonstrated its superiority. We also validated mmAAVI's ability for cell state knowledge transfer, achieving balanced accuracies of 0.82 and 0.97 with less 1% labeled cells between batches with completely different omics. The full package is available at https://github.com/luyiyun/mmAAVI.

ScDFN: enhancing single-cell RNA-seq clustering with deep fusion networks.

Single-cell ribonucleic acid sequencing (scRNA-seq) technology can be used to perform high-resolution analysis of the transcriptomes of individual cells. Therefore, its application has gained popularity for accurately analyzing the ever-increasing content of heterogeneous single-cell datasets. Central to interpreting scRNA-seq data is the clustering of cells to decipher transcriptomic diversity and infer cell behavior patterns. However, its complexity necessitates the application of advanced methodologies capable of resolving the inherent heterogeneity and limited gene expression characteristics of single-cell data. Herein, we introduce a novel deep learning-based algorithm for single-cell clustering, designated scDFN, which can significantly enhance the clustering of scRNA-seq data through a fusion network strategy. The scDFN algorithm applies a dual mechanism involving an autoencoder to extract attribute information and an improved graph autoencoder to capture topological nuances, integrated via a cross-network information fusion mechanism complemented by a triple self-supervision strategy. This fusion is optimized through a holistic consideration of four distinct loss functions. A comparative analysis with five leading scRNA-seq clustering methodologies across multiple datasets revealed the superiority of scDFN, as determined by better the Normalized Mutual Information (NMI) and the Adjusted Rand Index (ARI) metrics. Additionally, scDFN demonstrated robust multi-cluster dataset performance and exceptional resilience to batch effects. Ablation studies highlighted the key roles of the autoencoder and the improved graph autoencoder components, along with the critical contribution of the four joint loss functions to the overall efficacy of the algorithm. Through these advancements, scDFN set a new benchmark in single-cell clustering and can be used as an effective tool for the nuanced analysis of single-cell transcriptomics.

Prenatal exposure to maternal smoking and adult lung cancer risk: a nested case-control study using peripheral blood leukocyte DNA methylation prediction of exposure.

A prior study reported no association between prenatal smoking methylation scores and adult lung cancer risk adjusting for methylation-predicted adult smoking, without considering maternal smoking trends by birth cohort. To address this gap, we examined the association between prenatal smoking methylation scores and adult lung cancer, independent of methylation-predicted adult packyears and by birth cohort, in a study nested in CLUE II. Included were 208 incident lung cancer cases ascertained by cancer registry linkage and 208 controls matched on age, sex, and smoking. DNA methylation was measured in prediagnostic blood. We calculated two prenatal smoking scores, using 19 (Score-19) and 15 (Score-15) previously identified CpGs and a methylation-predicted adult packyears score. Conditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) adjusting for adult packyears score and batch effects. Score-15 was positively associated with lung cancer (per standard deviation, OR = 1.40, 95% CI = 1.10-1.79, P-trend = .006), especially in the 1930-1938 birth cohort (OR = 3.43, 95% CI = 1.55-7.60, P-trend = .002). Score-19 was associated only in the 1930-1938 birth cohort (OR = 2.12, 95% CI = 1.15-3.91). Participants with both prenatal scores below the median (vs all other combinations) had lower risk (OR = 0.44, 95% CI = 0.27-0.72), especially in the 1930-1938 birth cohort (OR = 0.16, 95% CI = 0.04-0.62). Among ever smokers, participants with higher prenatal smoking scores had higher risk, irrespective of adult packyears (low: OR = 2.81, 95% CI = 1.38-5.72, high: OR = 2.67, 95% CI = 1.03-6.95). This prospective study suggests a positive association between prenatal smoking exposure and adult lung cancer risk, especially in the 1930-1938 birth cohort, independent of active smoking. Future studies with multiple birth cohorts are needed.