Batch Effect Correction Research Articles

Exploring tumor endothelial cells heterogeneity in hepatocellular carcinoma: insights from single-cell sequencing and pseudotime analysis.

This study aimed to explore the heterogeneity of tumor endothelial cells (TECs) in hepatocellular carcinoma (HCC) and their role in tumor progression, with the goal of identifying new therapeutic targets and strategies to improve patient prognosis. Single-cell RNA sequencing data from nine primary liver cancer samples were analyzed, obtained from the Gene Expression Omnibus (GEO) database. Data preprocessing, normalization, dimensionality reduction, and batch effect correction were performed based on the Seurat package. HCC cell types were identified using uniform manifold approximation and projection (UMAP) and cluster analysis, and the different cell types were annotated using the CellMarker database. Pseudotime trajectory analysis was conducted with Monocle to explore the differentiation trajectory of TECs. MAPK signaling pathway activity and copy number variations (CNV) in TECs were analyzed in conjunction with data from The Cancer Genome Atlas (TCGA), the trans-well and wound healing assay was used for cell invasion and migration activity assessment. Two subgroups of TECs (TECs 1 and TECs 2) were identified, exhibiting distinct functional activities and signaling pathways. Specifically, TECs 1 may be involved in tumor cell proliferation and inflammatory responses, whereas TECs 2 is not only involved in cell proliferation pathways, but also enriched in pathways such as metabolic synthesis. Pseudotime analysis revealed dynamic changes in TECs subgroups during HCC progression, correlating specific gene expressions (such as PDGFRB, PGF, JUN, and NR4A1). Subsequently, the JUN gene was predicted by performing binding sites and was shown to act as a transcription factor that may regulate the expression of the PGF gene. CNV analysis highlighted key genes and pathways in TECs that might influence HCC progression, and the PGF as key regulatory factor mediated cell proliferation and migration. The study revealed the heterogeneity of TECs in HCC and their potential roles in tumor progression, offering new perspectives and potential therapeutic targets for HCC molecular mechanisms. The findings emphasize the importance of further exploring TECs heterogeneity for understanding HCC pathogenesis and developing personalized treatment strategies.

Batch effects correction in scRNA-seq based on biological-noise decoupling autoencoder and central-cross loss

Technical or biologically irrelevant differences caused by different experiments, times, or sequencing platforms can generate batch effects that mask the true biological information. Therefore, batch effects are typically removed when analyzing single-cell RNA sequencing (scRNA-seq) datasets for downstream tasks. Existing batch correction methods usually mitigate batch effects by reducing the data from different batches to a lower dimensional space before clustering, potentially leading to the loss of rare cell types. To address this problem, we introduce a novel single-cell data batch effect correction model using Biological-noise Decoupling Autoencoder (BDA) and Central-cross Loss termed BDACL. The model initially reconstructs raw data using an auto-encoder and conducts preliminary clustering. We then construct a similarity matrix and a hierarchical clustering tree to delineate relationships within and between different batches. Finally, we introduce a Central-cross Loss (CL). This loss leverages cross-entropy loss to prompt the model to better distinguish between different cluster labels. Additionally, it employs the Central Loss to encourage samples to form more compact clusters in the embedding space, thereby enhancing the consistency and interpretability of clustering results to mitigate differences between different batches. The primary innovation of this model lies in reconstructing data with an auto-encoder and gradually merging smaller clusters into larger ones using a hierarchical clustering tree. By using reallocated cluster labels as training labels and employing the Central-cross Loss, the model effectively eliminates batch effects in an unsupervised manner. Compared to current methods, BDACL can mitigate batch effects without losing rare cell types.

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.

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.

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.

A graph self-supervised residual learning framework for domain identification and data integration of spatial transcriptomics

Spatial transcriptomics (ST) technologies allow for comprehensive characterization of gene expression patterns in the context of tissue microenvironment. However, accurately identifying domains with spatial coherence in both gene expression and histology in situ and effectively integrating data from multi-sample remains challenging. Here, we propose ResST, a graph self-supervised residual learning model based on graph neural network and Margin Disparity Discrepancy (MDD) theory. ResST aggregates gene expression, biological effects, spatial location, and morphological information to capture nonlinear relationships between a cell and surrounding cells for spatial domain identification. Also, ResST integrates multiple ST datasets and aligns latent embeddings based on MDD theory for correcting batch effects. Results show that ResST identifies continuous spatial domains at a finer scale in ten ST datasets acquired with different technologies. Moreover, ResST efficiently integrated data from multiple tissue sections vertically or horizontally while correcting batch effects. Overall, ResST demonstrates exceptional performance in analyzing ST datasets.

ImmCellTyper facilitates systematic mass cytometry data analysis for deep immune profiling.

Mass cytometry is a cutting-edge high-dimensional technology for profiling marker expression at the single-cell level, advancing clinical research in immune monitoring. Nevertheless, the vast data generated by cytometry by time-of-flight (CyTOF) poses a significant analytical challenge. To address this, we describe ImmCellTyper (https://github.com/JingAnyaSun/ImmCellTyper), a novel toolkit for CyTOF data analysis. This framework incorporates BinaryClust, an in-house developed semi-supervised clustering tool that automatically identifies main cell types. BinaryClust outperforms existing clustering tools in accuracy and speed, as shown in benchmarks with two datasets of approximately 4 million cells, matching the precision of manual gating by human experts. Furthermore, ImmCellTyper offers various visualisation and analytical tools, spanning from quality control to differential analysis, tailored to users' specific needs for a comprehensive CyTOF data analysis solution. The workflow includes five key steps: (1) batch effect evaluation and correction, (2) data quality control and pre-processing, (3) main cell lineage characterisation and quantification, (4) in-depth investigation of specific cell types; and (5) differential analysis of cell abundance and functional marker expression across study groups. Overall, ImmCellTyper combines expert biological knowledge in a semi-supervised approach to accurately deconvolute well-defined main cell lineages, while maintaining the potential of unsupervised methods to discover novel cell subsets, thus facilitating high-dimensional immune profiling.

Higher-order correction of persistent batch effects in correlation networks.

Systems biology analyses often use correlations in gene expression profiles to infer co-expression networks that are then used as input for gene regulatory network inference or to identify functional modules of co-expressed or putatively co-regulated genes. While systematic biases, including batch effects, are known to induce spurious associations and confound differential gene expression analyses (DE), the impact of batch effects on gene co-expression has not been fully explored. Methods have been developed to adjust expression values, ensuring conditional independence of mean and variance from batch or other covariates for each gene, resulting in improved fidelity of DE analysis. However, such adjustments do not address the potential for spurious differential co-expression (DC) between groups. Consequently, uncorrected, artifactual DC can skew the correlation structure, leading to the identification of false, non-biological associations, even when the input data are corrected using standard batch correction. In this work, we demonstrate the persistence of confounders in covariance after standard batch correction using synthetic and real-world gene expression data examples. We then introduce Co-expression Batch Reduction Adjustment (COBRA), a method for computing a batch-corrected gene co-expression matrix based on estimating a conditional covariance matrix. COBRA estimates a reduced set of parameters expressing the co-expression matrix as a function of the sample covariates, allowing control for continuous and categorical covariates. COBRA is computationally efficient, leveraging the inherently modular structure of genomic data to estimate accurate gene regulatory associations and facilitate functional analysis for high-dimensional genomic data. COBRA is available under the GLP3 open source license in R and Python in netZoo (https://netzoo.github.io).

Computational Analysis of Marker Genes in Alzheimer’s Disease across Multiple Brain Regions

Alzheimer’s disease (AD) is the most common cause of neurodegenerative dementia in the elderly, which is characterized by progressive cognitive impairment. Herein, we undertake a sophisticated computational analysis by integrating single-cell RNA sequencing (scRNA-seq) data from multiple brain regions significantly affected by the disease, including the entorhinal cortex, prefrontal cortex, superior frontal gyrus, and superior parietal lobe. Our pipeline combines datasets derived from the aforementioned tissues into a unified analysis framework, facilitating cross-regional comparisons to provide a holistic view of the impact of the disease on the cellular and molecular landscape of the brain. We employed advanced computational techniques such as batch effect correction, normalization, dimensionality reduction, clustering, and visualization to explore cellular heterogeneity and gene expression patterns across these regions. Our findings suggest that enabling the integration of data from multiple batches can significantly enhance our understanding of AD complexity, thereby identifying key molecular targets for potential therapeutic intervention. This study established a precedent for future research by demonstrating how existing data can be reanalysed in a coherent manner to elucidate the systemic nature of the disease and inform the development of more effective diagnostic tools and targeted therapies.

MRMQuant: Automated MRM Data Quantitation for Large-Scale Targeted Metabolomics Analysis.

Multiple reaction monitoring (MRM) is a powerful and popular technique used for metabolite quantification in targeted metabolomics. Accurate and consistent quantitation of metabolites from the MRM data is essential for subsequent analyses. Here, we developed an automated tool, MRMQuant, for targeted metabolomic quantitation using high-throughput liquid chromatography-tandem mass spectrometry MRM data to provide users with an easy-to-use tool for accurate MRM data quantitation with minimal human intervention. This tool has many user-friendly functions and features to inspect and correct the quantitation results as required. MRMQuant possesses the following features to ensure accurate quantitation: (1) dynamic signal smoothing, (2) automatic deconvolution of coeluted peaks, (3) absolute quantitation via standard curves and/or internal standards, (4) visualized inspection and correction, (5) corrections applicable to multiple samples, and (6) batch-effect correction.

COMSE: analysis of single-cell RNA-seq data using community detection-based feature selection

BackgroundSingle-cell RNA sequencing enables studying cells individually, yet high gene dimensions and low cell numbers challenge analysis. And only a subset of the genes detected are involved in the biological processes underlying cell-type specific functions.ResultIn this study, we present COMSE, an unsupervised feature selection framework using community detection to capture informative genes from scRNA-seq data. COMSE identified homogenous cell substates with high resolution, as demonstrated by distinguishing different cell cycle stages. Evaluations based on real and simulated scRNA-seq datasets showed COMSE outperformed methods even with high dropout rates in cell clustering assignment. We also demonstrate that by identifying communities of genes associated with batch effects, COMSE parses signals reflecting biological difference from noise arising due to differences in sequencing protocols, thereby enabling integrated analysis of scRNA-seq datasets of different sources.ConclusionsCOMSE provides an efficient unsupervised framework that selects highly informative genes in scRNA-seq data improving cell sub-states identification and cell clustering. It identifies gene subsets that reveal biological and technical heterogeneity, supporting applications like batch effect correction and pathway analysis. It also provides robust results for bulk RNA-seq data analysis.

ScMaui: a widely applicable deep learning framework for single-cell multiomics integration in the presence of batch effects and missing data

The recent advances in high-throughput single-cell sequencing have created an urgent demand for computational models which can address the high complexity of single-cell multiomics data. Meticulous single-cell multiomics integration models are required to avoid biases towards a specific modality and overcome sparsity. Batch effects obfuscating biological signals must also be taken into account. Here, we introduce a new single-cell multiomics integration model, Single-cell Multiomics Autoencoder Integration (scMaui) based on variational product-of-experts autoencoders and adversarial learning. scMaui calculates a joint representation of multiple marginal distributions based on a product-of-experts approach which is especially effective for missing values in the modalities. Furthermore, it overcomes limitations seen in previous VAE-based integration methods with regard to batch effect correction and restricted applicable assays. It handles multiple batch effects independently accepting both discrete and continuous values, as well as provides varied reconstruction loss functions to cover all possible assays and preprocessing pipelines. We demonstrate that scMaui achieves superior performance in many tasks compared to other methods. Further downstream analyses also demonstrate its potential in identifying relations between assays and discovering hidden subpopulations.

NDMNN: A novel deep residual network based MNN method to remove batch effects from scRNA-seq data.

The rapid development of single-cell RNA sequencing (scRNA-seq) technology has generated vast amounts of data. However, these data often exhibit batch effects due to various factors such as different time points, experimental personnel, and instruments used, which can obscure the biological differences in the data itself. Based on the characteristics of scRNA-seq data, we designed a dense deep residual network model, referred to as NDnetwork. Subsequently, we combined the NDnetwork model with the MNN method to correct batch effects in scRNA-seq data, and named it the NDMNN method. Comprehensive experimental results demonstrate that the NDMNN method outperforms existing commonly used methods for correcting batch effects in scRNA-seq data. As the scale of single-cell sequencing continues to expand, we believe that NDMNN will be a valuable tool for researchers in the biological community for correcting batch effects in their studies. The source code and experimental results of the NDMNN method can be found at https://github.com/mustang-hub/NDMNN.

Comparing preprocessing strategies for 3D-Gene microarray data of extracellular vesicle-derived miRNAs

BackgroundExtracellular vesicle-derived (EV)-miRNAs have potential to serve as biomarkers for the diagnosis of various diseases. miRNA microarrays are widely used to quantify circulating EV-miRNA levels, and the preprocessing of miRNA microarray data is critical for analytical accuracy and reliability. Thus, although microarray data have been used in various studies, the effects of preprocessing have not been studied for Toray’s 3D-Gene chip, a widely used measurement method. We aimed to evaluate batch effect, missing value imputation accuracy, and the influence of preprocessing on measured values in 18 different preprocessing pipelines for EV-miRNA microarray data from two cohorts with amyotrophic lateral sclerosis using 3D-Gene technology.ResultsEighteen different pipelines with different types and orders of missing value completion and normalization were used to preprocess the 3D-Gene microarray EV-miRNA data. Notable results were suppressed in the batch effects in all pipelines using the batch effect correction method ComBat. Furthermore, pipelines utilizing missForest for missing value imputation showed high agreement with measured values. In contrast, imputation using constant values for missing data exhibited low agreement.ConclusionsThis study highlights the importance of selecting the appropriate preprocessing strategy for EV-miRNA microarray data when using 3D-Gene technology. These findings emphasize the importance of validating preprocessing approaches, particularly in the context of batch effect correction and missing value imputation, for reliably analyzing data in biomarker discovery and disease research.

A deep learning paradigm for medical imaging data

We present a novel deep learning framework to address several challenges in medical imaging analysis. Especially for functional magnetic resonance imaging (fMRI) data, we develop a statistical sampling algorithm to augment small data for deep learning modeling. Also, the ComBat harmonization is utilized for multi-source batch effect correction. Finally, we implement the prediction uncertainty in deep learning through a dropout approximation of the deep Gaussian process. The proposed approach offers effective statistical tools to solve persistent problems in deep learning modeling for medical imaging data. To assess the performance of the proposed approach and demonstrate its effectiveness for various types of data, we apply it to autism fMRI and Pneumonia imaging datasets.

Data Harmonization Guidelines to Combine Multi-platform Genomic Data from Admixed Populations and Boost Power in Genome-Wide Association Studies.

Data harmonization involves combining data from multiple independent sources and processing the data to produce one uniform dataset. Merging separate genotypes or whole-genome sequencing datasets has been proposed as a strategy to increase the statistical power of association tests by increasing the effective sample size. However, data harmonization is not a widely adopted strategy due to the difficulties with merging data (including confounding produced by batch effects and population stratification). Detailed data harmonization protocols are scarce and are often conflicting. Moreover, data harmonization protocols that accommodate samples of admixed ancestry are practically non-existent. Existing data harmonization procedures must be modified to ensure the heterogeneous ancestry of admixed individuals is incorporated into additional downstream analyses without confounding results. Here, we propose a set of guidelines for merging multi-platform genetic data from admixed samples that can be adopted by any investigator with elementary bioinformatics experience. We have applied these guidelines to aggregate 1544 tuberculosis (TB) case-control samples from six separate in-house datasets and conducted a genome-wide association study (GWAS) of TB susceptibility. The GWAS performed on the merged dataset had improved power over analyzing the datasets individually and produced summary statistics free from bias introduced by batch effects and population stratification. © 2024 Wiley Periodicals LLC. Basic Protocol 1: Processing separate datasets comprising array genotype data Alternate Protocol 1: Processing separate datasets comprising array genotype and whole-genome sequencing data Alternate Protocol 2: Performing imputation using a local reference panel Basic Protocol 2: Merging separate datasets Basic Protocol 3: Ancestry inference using ADMIXTURE and RFMix Basic Protocol 4: Batch effect correction using pseudo-case-control comparisons.

Deep Batch Integration and Denoise of Single-Cell RNA-Seq Data.

Numerous single-cell transcriptomic datasets from identical tissues or cell lines are generated from different laboratories or single-cell RNA sequencing (scRNA-seq) protocols. The denoising of these datasets to eliminate batch effects is crucial for data integration, ensuring accurate interpretation and comprehensive analysis of biological questions. Although many scRNA-seq data integration methods exist, most are inefficient and/or not conducive to downstream analysis. Here, DeepBID, a novel deep learning-based method for batch effect correction, non-linear dimensionality reduction, embedding, and cell clustering concurrently, is introduced. DeepBID utilizes a negative binomial-based autoencoder with dual Kullback-Leibler divergence loss functions, aligning cell points from different batches within a consistent low-dimensional latent space and progressively mitigating batch effects through iterative clustering. Extensive validation on multiple-batch scRNA-seq datasets demonstrates that DeepBID surpasses existing tools in removing batch effects and achieving superior clustering accuracy. When integrating multiple scRNA-seq datasets from patients with Alzheimer's disease, DeepBID significantly improves cell clustering, effectively annotating unidentified cells, and detecting cell-specific differentially expressed genes.

Integrative Single-Cell Analysis of Cardiomyopathy Identifies Differences in Cell Stemness and Transcriptional Regulatory Networks among Fibroblast Subpopulations.

Cardiomyopathy encompasses a broad spectrum of diseases affecting myocardial tissue, characterized clinically by abnormalities in cardiac structure, heart failure, and/or arrhythmias. Clinically heterogeneous, major types include dilated cardiomyopathy (DCM), hypertrophic cardiomyopathy (HCM), restrictive cardiomyopathy (RM), ischemic cardiomyopathy (ICM), among which DCM is more prevalent, while ICM exhibits higher incidence and mortality rates. Myocardial injury during cardiomyopathy progression may lead to myocardial fibrosis. Failure to intervene early and inhibit the process of myocardial fibrosis may culminate in heart failure. Cardiac fibroblasts constitute crucial cellular components determining the extent and quality of myocardial fibrosis, with various subpopulations exerting diverse roles in cardiomyopathy progression. Despite this, understanding of the cellular plasticity and transcriptional regulatory networks of cardiac fibroblasts in cardiomyopathy remains limited. Therefore, in this study, we conducted comprehensive single-cell analysis of cardiac fibroblasts in cardiomyopathy to explore differences in cellular plasticity and transcriptional regulatory networks among fibroblast subpopulations, with the aim of providing as many useful references as possible for the diagnosis, prognosis, and treatment of cardiomyopathy. Cells with mitochondrial gene expression comprising >20% of total expressed genes were excluded. Differential expression genes (DEGs) and stemness genes within cardiac fibroblast subpopulations were subjected to Gene Ontology (GO) analysis of biological processes (BP) and AUCell analysis. Monocle software was employed to analyze the pseudo-temporal trajectory of cardiac fibroblasts in cardiomyopathy. Additionally, the Python package SCENIC was utilized to assess enrichment of transcription factors and activity of regulators within cardiac fibroblast subpopulations in cardiomyopathy. Following batch effect correction, 179,927 cells were clustered into 32 clusters, designated as T_NK cells, endothelial cells, myeloid cells, fibroblasts, pericytes, SMCs, CMs, proliferating cells, EndoCs, and EPCs. Among them, 8148 fibroblasts were further subdivided into 4 subpopulations, namely C0 THBS4+ Fibroblasts, C1 LINC01133+ Fibroblasts, C2 FGF7+ Fibroblasts, and C3 AGT + Fibroblasts. Results from GO_BP and AUCell analyses suggest that C3 AGT + Fibroblasts may be associated with immune response activation, protein transport, and myocardial contractile function, correlating with disease progression in cardiomyopathy. Transcription factor enrichment analysis indicates that FOS is the most significant TF in C3 AGT + Fibroblasts, also associated with the M1 module, possibly implicated in protein hydrolysis, intracellular DNA replication, and cell proliferation. Moreover, correlation analysis of transcriptional regulatory activity between fibroblast subpopulations reveals a more pronounced heterogeneity within C3 AGT + Fibroblasts in cardiomyopathy. C3 AGT + Fibroblasts exhibit increased sensitivity towards adverse outcomes in cardiomyopathy, such as myocardial fibrosis and impaired cardiac contractile function, compared to other cardiac fibroblast subpopulations. The differential cellular plasticity and transcriptional regulatory activity between C3 AGT + Fibroblasts and other subgroups offer new perspectives for targeting fibroblast subpopulation activity to treat cardiomyopathy. Additionally, stemness genes EPAS1 and MYC, along with the regulator FOS, may play roles in modulating the biological processes of cardiac fibroblasts in cardiomyopathy.

A comparison of RNA-Seq data preprocessing pipelines for transcriptomic predictions across independent studies

BackgroundRNA sequencing combined with machine learning techniques has provided a modern approach to the molecular classification of cancer. Class predictors, reflecting the disease class, can be constructed for known tissue types using the gene expression measurements extracted from cancer patients. One challenge of current cancer predictors is that they often have suboptimal performance estimates when integrating molecular datasets generated from different labs. Often, the quality of the data is variable, procured differently, and contains unwanted noise hampering the ability of a predictive model to extract useful information. Data preprocessing methods can be applied in attempts to reduce these systematic variations and harmonize the datasets before they are used to build a machine learning model for resolving tissue of origins.ResultsWe aimed to investigate the impact of data preprocessing steps—focusing on normalization, batch effect correction, and data scaling—through trial and comparison. Our goal was to improve the cross-study predictions of tissue of origin for common cancers on large-scale RNA-Seq datasets derived from thousands of patients and over a dozen tumor types. The results showed that the choice of data preprocessing operations affected the performance of the associated classifier models constructed for tissue of origin predictions in cancer.ConclusionBy using TCGA as a training set and applying data preprocessing methods, we demonstrated that batch effect correction improved performance measured by weighted F1-score in resolving tissue of origin against an independent GTEx test dataset. On the other hand, the use of data preprocessing operations worsened classification performance when the independent test dataset was aggregated from separate studies in ICGC and GEO. Therefore, based on our findings with these publicly available large-scale RNA-Seq datasets, the application of data preprocessing techniques to a machine learning pipeline is not always appropriate.