limma statistics

To access cutting-edge analytics on consensus tools, life science contexts and associated fields, you will need to subscribe to our premium service.


Citations per year

Citations chart

Popular tool citations

chevron_left Gene set enrichment analysis Normalization Differential expression detection Differential expression Differential expression detection Batch effect correction chevron_right
Popular tools chart

Tool usage distribution map

Tool usage distribution map

Associated diseases

Associated diseases


To access compelling stats and trends, optimize your time and resources and pinpoint new correlations, you will need to subscribe to our premium service.


limma specifications


Unique identifier OMICS_00769
Name limma
Alternative names Linear Models for Microarray Data, r-bioc-limma
Software type Package/Module
Interface Command line interface
Restrictions to use None
Operating system Unix/Linux, Mac OS, Windows
Programming languages R
License GNU General Public License version 2.0
Computer skills Advanced
Version 3.36.5
Stability Stable
AnnotationDbi, methods, stats, gplots, graphics, Biobase, splines, utils, grDevices, locfit, affy, MASS, illuminaio, ellipse, R(>=2.3.0), BiasedUrn, GO.db,, statmod(>=1.2.2), vsn
Maintained Yes





Add your version



  • person_outline Gordon Smyth <>

Publications for Linear Models for Microarray Data

limma in pipelines

PMCID: 5753908
PMID: 29291611
DOI: 10.5056/jnm17021

[…] using agilent’s feature extraction software (version 11.01.1) using protocol ge1_1100_jul11 and the 60k grid file 039494_d_f_20120628., background-corrected intensities were normalized by the “limma” package in r. differentially expressed genes between groups were obtained by combining the lists of features with p-values < 0.05 from the “limma” linear model t test, the mann-whitney u […]

PMCID: 5770019
PMID: 29338044
DOI: 10.1371/journal.pone.0190817

[…] grouped based on known kyoto encyclopedia of genes and genomes (kegg) biological interactions and pathways. we used the r bioconductor package gage [] to examine the log 2 fold-changes reported by limma to determine if genes expressed in each contrast, regardless of statistical significance, were up or down regulated in comparison to global background log 2 fold-changes as well as tests […]

PMCID: 5774019
PMID: 29337789
DOI: 10.1097/CCM.0000000000002839

[…] review board (protocol 29798)., gene expression datasets that used affymetrix arrays were gcrma normalized using r package affy, and other arrays were quantile normalized using r package limma () if not already normalized. all microarray data were log-2 transformed, and probes were summarized to genes within datasets using a fixed-effect inverse variance model., multicohort analysis […]

PMCID: 5776392
PMID: 29387063
DOI: 10.3389/fimmu.2017.02004

[…] language (version 3.2.3). the dataset was analyzed using the oligo package from bioconductor. multiple probes were collapsed to single gene using the average expression (avereps function). using the limma package, a linear model was fitted for the identification of differentially expressed genes. genes with fdr ≤ 0.05 and log2fc ≥ +1.0 (upregulated) or log2fc ≤ −1.0 (downregulated) […]

PMCID: 5786503
PMID: 29275859
DOI: 10.1016/j.cell.2017.11.042

[…] expression level of ddx3y and xist genes. genes that were significantly differentially expressed between microglia from early versus late fetuses, or female versus male fetuses, were selected using limma and had adjusted p values less than 0.05. human gene ids and mouse gene ids were converted to homologene group id (hid) based on ncbi homologene. converted human and murine microarray gene […]

To access a full list of citations, you will need to upgrade to our premium service.

limma in publications

PMCID: 5958058
PMID: 29773832
DOI: 10.1038/s41467-018-04383-6

[…] classifiers were built for each dataset describing a classification using an approach described in previous works, . briefly, after gene-wise centering, the 1000 most differentially expressed genes (limma) or differentially methylated cpg (student’s t-test), were used to build centroids of each subtype. gene expression profiles of samples to test were correlated (pearson’s correlation) […]

PMCID: 5954024
PMID: 29765119
DOI: 10.1038/s41598-018-25911-w

[…] **mean number of total metastatic lns., as noted above, left-over osna samples from 8 patients, 4 cases each from groups a and b, were analyzed with mass spectrometry. among 1228 proteins examined, limma analysis revealed 83 proteins which differed significantly in their expression levels between the two groups. thirty-eight proteins were highly expressed and 45 showed lower expressions […]

PMCID: 5954140
PMID: 29765046
DOI: 10.1038/s41467-018-04244-2

[…] accordance with the manufacturer’s protocols. after the bead chips were scanned with an illumina beadarray reader, the microarray data were normalized using the quantile normalization method in the linear models for microarray data package in the r language environment. the expression level of each gene was then log2 transformed before further analysis. the microarray data are available […]

PMCID: 5953988
PMID: 29765129
DOI: 10.1038/s41598-018-25762-5

[…] from the dataset. statistical analyses were performed using the statistical programming environment implemented in carmaweb. gene wise testing for differential expression was done employing the (limma) t-test and benjamini-hochberg multiple testing correction (fdr < 10%). sets of significantly regulated mirnas were filtered for present calls in at least 60% of the samples in at least one […]

PMCID: 5951826
PMID: 29760449
DOI: 10.1038/s41598-018-24982-z

[…] biological replicates were performed for each sirna experiment (both aurka and non-targeting). pre-processing and differential expression analysis (dea) was done using the r-bioconductor packages limma and sva-combat. in short, conventional background correction was applied, followed by within-array normalization using loess. subsequently, batch-effect removal was applied for the replicate […]

To access a full list of publications, you will need to upgrade to our premium service.

limma institution(s)
Bioinformatics Division, The Walter and Eliza Hall Institute of Medical Research, Parkville, VIC, Australia
limma funding source(s)
National Health and Medical Research Council Project Grant [1050661 and 1023454]; NHMRC Program Grant [1054618]; Victorian State Government Operational Infrastructure Support and Australian Government NHMRC IRIISS

limma review

star_border star_border star_border star_border star_border
star star star star star


star_border star_border star_border star_border star_border
star star star star star
limma is a tool for reading, pre-processing, manipulate, quality control and analyse microarray data from a variety of different technologies. In my opinion, it could easily be considered the state-of-the-art tool for analysis of these kind of data. Linear modelling extended by limma provides a powerful way to analyse simple or complex experiments, allowing full control over covariates and time-series designs. Also, the team behind the tool made great effort in creating a tool easily accessible to non bio-statisticians, which is a plus in a lot of ways. Besides, I should also point out that limma has the capability to analyse RNA-seq experiments as well. Finally, users should definitely check the comprehensive (and well-written) limma User Guide, a document that will provide all the basics to get things rolling with the package.