Mitigating the adverse impact of batch effects in sample pattern detection

Fei, Teng; Zhang, Tengjiao; Shi, Weixing; Yu, Tianwei

doi:10.1093/bioinformatics/bty117

Cited by 18 publications

(31 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For mixed transcriptome samples, 30% of reads mapped to the S. gordonii genome and 60% of reads mapped to the F. nucleatum genome (Table ). Samples from mixed and monoculture samples showed a similar distribution of per‐gene read counts per sample, as visualized by box plots (Figure ), indicating that the distributions of data were quantitatively comparable between mixed coaggregate and monoculture samples and no batch effects …”

Section: Resultsmentioning

confidence: 69%

Transcriptional responses of Streptococcus gordonii and Fusobacterium nucleatum to coaggregation

Mutha

Mohammed

Krasnogor

et al. 2018

Molecular Oral Microbiology

View full text Add to dashboard Cite

Cell‐cell interactions between genetically distinct bacteria, known as coaggregation, are important for the formation of mixed‐species biofilms such as dental plaque. Interactions lead to gene regulation in the partner organisms that may be critical for adaptation and survival in mixed‐species biofilms. Here, gene regulation responses to coaggregation between Streptococcus gordonii and Fusobacterium nucleatum were studied using dual RNA‐Seq. Initially, S. gordonii was shown to coaggregate strongly with F. nucleatum in buffer or human saliva. Using confocal laser scanning microscopy and transmission electron microscopy, cells of different species were shown to be evenly distributed throughout the coaggregate and were closely associated with one another. This distribution was confirmed by serial block face sectioning scanning electron microscopy, which provided high resolution three‐dimensional images of coaggregates. Cell‐cell sensing responses were analysed 30 minutes after inducing coaggregation in human saliva. By comparison with monocultures, 16 genes were regulated following coaggregation in F. nucleatum whereas 119 genes were regulated in S. gordonii. In both species, genes involved in amino acid and carbohydrate metabolism were strongly affected by coaggregation. In particular, one 8‐gene operon in F. nucleatum encoding sialic acid uptake and catabolism was up‐regulated 2‐ to 5‐fold following coaggregation. In S. gordonii, a gene cluster encoding functions for phosphotransferase system‐mediated uptake of lactose and galactose was down‐regulated up to 3‐fold in response to coaggregation. The genes identified in this study may play key roles in the development of mixed‐species communities and represent potential targets for approaches to control dental plaque accumulation.

show abstract

Section: Resultsmentioning

confidence: 69%

Transcriptional responses of Streptococcus gordonii and Fusobacterium nucleatum to coaggregation

Mutha

Mohammed

Krasnogor

et al. 2018

Molecular Oral Microbiology

View full text Add to dashboard Cite

show abstract

“…Without batch effect correction, the samples from the same species clustered together, as the tissues from different species were measured in different batches. Several re-analyses have shown that the samples would be clustered by tissues instead of species after proper batch effect correction (Gilad and Mizrahi-Man, 2015; Sudmant et al, 2015; Fei et al, 2018).…”

Section: Resultsmentioning

confidence: 99%

“…For scBatch, the restoring of sample patterns was based on our previously published work of sample correlation matrix correction (Fei et al, 2018). Thus the results here on the mouse/human tissue data only demonstrated that scBatch could truly adjust the count matrix such that the resulting sample correlation matrix approaches that generated by Fei et al (2018). In the next sections, we demonstrate the utility of scBatch in DE gene detection on single cell RNAseq data.…”

Section: Resultsmentioning

confidence: 99%

“…ComBat and the control gene methods are based on regression models, while more recent methods proposed different strategies that allow for more complex batch effect mechanisms. To achieve better clustering performance, Fei et al (2018) developed a non-parametric approach, named QuantNorm, to correct sample distance matrix by quantile normalization; Haghverdi et al (2018) utilized the mutual nearest neighbor relationships among samples from different batches to establish the MNN correction scheme. As observed from the results in Fei et al (2018) and Haghverdi et al (2018), current methods have reached reasonable performances in sample pattern detection, such as finding clusters or conducting dimension reduction.…”

Section: Introductionmentioning

confidence: 99%

“…However, DE detection appears not to be the emphasis of recent methods development. Chen and Zhou (2017) only evaluated the clustering performances, while QuantNorm (Fei et al, 2018) only returns corrected distance matrices and does not support DE tests. Although Haghverdi et al (2018) conducted DE tests, the user manual (https://bioconductor.org/packages/3.8/workflows/vignettes/simpleSingleCell/inst/doc/work-5-mnn.html) of the corresponding bioconductor function, mnnCorrect, recommends not using the corrected count matrix for DE analysis with considerations on manipulated data scales and mean-variance relationship.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Batch Effect Correction of RNA-seq Data through Sample Distance Matrix Adjustment

Fei

2019

Preprint

Self Cite

View full text Add to dashboard Cite

8Batch effect is a frequent challenge in deep sequencing data analysis that can lead to misleading 9 conclusions. We present scBatch, a numerical algorithm that conducts batch effect correction on the 10 count matrix of RNA sequencing (RNA-seq) data. Different from traditional methods, scBatch starts 11 with establishing an ideal correction of the sample distance matrix that effectively reflect the underlying 12 biological subgroups, without considering the actual correction of the raw count matrix itself. It then 13 seeks an optimal linear transformation of the count matrix to approximate the established sample pattern. 14 The benefit of such an approach is the final result is not restricted by assumptions on the mechanism of 15 the batch effect. As a result, the method yields good clustering and gene differential expression (DE) 16 results. We compared the new method, scBatch, with leading batch effect removal methods ComBat 17 and mnnCorrect on simulated data, real bulk RNA-seq data, and real single-cell RNA-seq data. The 18 comparisons demonstrated that scBatch achieved better sample clustering and DE gene detection results.In the recent decade, RNA sequencing (RNA-seq) has become a major tool for transcriptomics. Due 21 to the limitation of sequencing technology and sample preparations, technical variations exist among 22 reads from different batches of experiments. These unwanted technical variations, or batch effects, can 23 lead to misleading scientific findings in downstream data analysis (Hicks et al., 2017). Typically, batch 24 effects can alter the sample patterns, causing false interpretations about cell lineage and heterogeneity. 25 If the goal is to detect differential expression (DE) genes, the analysis can suffer loss of statistical power 26 and/or bias. 27While the severity of batch effects varies in different datasets, batch effect corrections were shown to 28 be effective in general. For instance, batch effect correction on the ENCODE human and mouse tissues 29 bulk RNA-seq data (Lin et al., 2014), where the batch effects were intense, obtained largely different and 30 more sensible tissue clustering results compared to before correction (Gilad and Mizrahi-Man, 2015). 31In other datasets, batch effects are often more subtle. In such cases, although the true biological pattern 32 is maintained to some extent, weak to moderate batch effects can still be observed. Hicks et al. (2017) 33 discussed the coexistence of biological signal and technical variation, which may still compromise the 34 downstream analysis. The correction of the batch effects can yield better clustering results (Fei et al., 35 2018) on data with weak to moderate batch effects that were unobvious from dimension reduction plots 36 (Usoskin et al., 2015; Muraro et al., 2016). These previous efforts argue for the inclusion of batch effect 37 corrections as a routine procedure in data preparation. 38Since the microarray era, efforts have been made to correct batch effects. Johnson et al. (2007) 39proposed an e...

show abstract

Assessment of a computerized quantitative quality control tool for whole slide images of kidney biopsies

Chen

Zee

Smith

et al. 2021

The Journal of Pathology

View full text Add to dashboard Cite

Inconsistencies in the preparation of histology slides and whole‐slide images (WSIs) may lead to challenges with subsequent image analysis and machine learning approaches for interrogating the WSI. These variabilities are especially pronounced in multicenter cohorts, where batch effects (i.e. systematic technical artifacts unrelated to biological variability) may introduce biases to machine learning algorithms. To date, manual quality control (QC) has been the de facto standard for dataset curation, but remains highly subjective and is too laborious in light of the increasing scale of tissue slide digitization efforts. This study aimed to evaluate a computer‐aided QC pipeline for facilitating a reproducible QC process of WSI datasets. An open source tool, HistoQC, was employed to identify image artifacts and compute quantitative metrics describing visual attributes of WSIs to the Nephrotic Syndrome Study Network (NEPTUNE) digital pathology repository. A comparison in inter‐reader concordance between HistoQC aided and unaided curation was performed to quantify improvements in curation reproducibility. HistoQC metrics were additionally employed to quantify the presence of batch effects within NEPTUNE WSIs. Of the 1814 WSIs (458 H&E, 470 PAS, 438 silver, 448 trichrome) from n = 512 cases considered in this study, approximately 9% (163) were identified as unsuitable for subsequent computational analysis. The concordance in the identification of these WSIs among computational pathologists rose from moderate (Gwet's AC1 range 0.43 to 0.59 across stains) to excellent (Gwet's AC1 range 0.79 to 0.93 across stains) agreement when aided by HistoQC. Furthermore, statistically significant batch effects (p < 0.001) in the NEPTUNE WSI dataset were discovered. Taken together, our findings strongly suggest that quantitative QC is a necessary step in the curation of digital pathology cohorts. © 2020 The Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.

show abstract

Mitigating the adverse impact of batch effects in sample pattern detection

Abstract: Supplementary data are available at Bioinformatics online.

Cited by 18 publications

References 21 publications

Transcriptional responses of Streptococcus gordonii and Fusobacterium nucleatum to coaggregation

Transcriptional responses of Streptococcus gordonii and Fusobacterium nucleatum to coaggregation

Batch Effect Correction of RNA-seq Data through Sample Distance Matrix Adjustment

Assessment of a computerized quantitative quality control tool for whole slide images of kidney biopsies

Contact Info

Product

Resources

About