About MPSBase

The database aims to help researchers to understand the pathophysiology of mucopolysaccharidoses, with up-to-date transcriptomic analyses. The project was developed at the Universidade Federal do Rio Grande do Sul - Brazil.

How to cite MPSBase?

Soares, L. D. F., Silva, G. C. V., Kubaski, F., Giugliani, R., & Matte, U. (2021). MPSBase: Comprehensive repository of differentially expressed genes for mucopolysaccharidoses. Molecular Genetics and Metabolism . Advance online publication. https://doi.org/10.1016/j.ymgme.2021.06.004

MPSBase development

The data present in this work are available in two public databases of gene expression studies, GEO datasets, and ArrayExpress. The thresholds used for differentially expressed genes (DEGs) filtering were p < 0.05 on FDR and |logFC| > 1.5. These analysis were conducted mainly with 4 R packages, accordingly to the platform of the experiment, described on the table.

Platform	Package	Normalization method
AB SOLiD	edgeR	calcNormFactors
Affymetrix exon array	oligo	rma
Affymetrix genomic array	affy	rma
Agilent	limma	backgroundCorrect(normexp.method = “rma”) normalizeBetweenArrays(method = “quantile”)
Illumina BeadChip	limma	backgroundCorrect(normexp.method = “rma”) normalizeBetweenArrays(method = “quantile”)
Illumina HiSeq	edgeR	calcNormFactors

Our database was implemented with MySQL and PHP. The logical model which describes its organization follows:

For gene ontology (GO) enrichment analysis, we used clusterProfiler and enrichKEGG packages, using p < 0.01 on BH correction as a threshold, and applying it only on the first 5000 top DEGs when the total number were more significant. To automate the Gene ontology enrichment for the 4 categories, we develop the autoGO R package. A one-step program to perform enrichment analysis, plotting the GO terms on multiple DEGs tables, and being flexible to species and an ID type. The autoGO package can be installed on the R environment with the commands above.

if (!requireNamespace("devtools", quietly = TRUE)){
                        install.packages("devtools")}
devtools::install_github("ldiass/MPSBase", subdir = "autoGO")

Exceptionally, we performed a quite different pipeline for the DEGs analysis on the C. lupus dataset. We choose to gather the genes using the Blast tool against the annotation of cDNA v.3.1 available on Ensembl due to the missing of a better annotation of the chip probes. Selecting only the hits with 100% of identity, we increased the number of assigned probes by 12%. The annotation proposed for Agilent-021193 Canine (V2) chip can be accessed here.