crossmeta
streamlines the cross-platform meta-analysis of
microarray data. For the analysis, you will need a list of Affymetrix, Illumina,
and/or Agilent GSE numbers from GEO. All 21
species in the current homologene build are supported.
See vignette for detailed usage.
crossmeta
is available through Bioconductor. To install the latest release from github:
install.packages('remotes')
remotes::install_github('alexvpickering/crossmeta')
# studies from GEO
gse_names <- c("GSE9601", "GSE15069")
# get raw data for specified studies
get_raw(gse_names)
# load and annotate raw data
esets <- load_raw(gse_names)
# perform differential expression analysis
anals <- diff_expr(esets)
# perform meta-analysis
es <- es_meta(anals)
A high quality meta-analysis is achieved by addressing the key issues in conducting a meta-analysis of microarray data (1):
Different labs process their raw data differently. These differences in data processing may lead to flawed conclusions upon meta-analysis.
crossmeta
starts with raw data, and uses a consistent processing pipeline
for all studies.
Related genes from different species are often referenced with different symbols. These differences can make it challenging to compare similar microarray experiments in diverse species.
crossmeta
maps probes to human gene symbols using homology relationships
established by HomoloGene.
Single probes can measure multiple genes. To incorporate these measurements appropriately, they are replaced with a new record for each gene.
Multiple probes can measure the same gene. Averaging these measurements is
inappropriate because measurement scales will vary with probe affinity.
crossmeta
selects the measurement with the highest inter-quartile range as
it is the least likely to occur by chance.
In addition to variables of interest, there are sources of signal due to factors that are unknown, unmeasured, or too complicated to capture through simple models. These factors can either hide true effects or introduce spurious ones.
crossmeta
discovers and accounts for these nuissance variables using
surrogate variable analysis.
Correctly specifying a model and contrast matrix can be challenging.
crossmeta
uses an attractive user interface that allows you to simply select
the samples you want to compare.
Paired samples (eg. the same subject before and after treatment) can also be selected using the same interface.
Differences in microarray platforms often lead to thousands of genes that are not measured in all studies. Most existing meta-analysis software requires that these genes with missing data are discarded.
crossmeta
extends the effect size meta-analysis method in GeneMeta
to
allow for genes that were not measured in all studies. Keep your data, find
more insights.
(1) Ramasamy A, Mondry A, Holmes CC, Altman DG (2008) Key Issues in Conducting a Meta-Analysis of Gene Expression Microarray Datasets. PLoS Med 5(9): e184. doi:10.1371/journal.pmed.0050184
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.