vignettes/fsgor.md

In DanWiebe/fsgor: Package for Fold-specific GO Terms Recognition

FSGOR package guide

Daniil Wiebe

2018-01-23

The aim of the package is to provide tools for identification of a set of genes with a similar degree of fold change associated with particular GO term (fold-change specific GO term) (Omelyanchuk et al., 2017). The package works with preprocessed experimental data on differential expression. Below the pipeline for RNA-seq experiment data is presented.

Upload differential expression data as a dataframe. The dataset in the example contains three columns 1) Gene ID 2) log fold-change value 3) p-value 4) adjusted p-value.

exp_path <- system.file("extdata", "rna_seq_example.txt", package = "fsgor")
rna_seq_data <- read.table(exp_path, stringsAsFactors = F, header = T)

Extract all genes under study as a vector

bg_genes <- rna_seq_data[, 1]

Divide initial dataframe into two with up-regulated and down-regulated differentially expressed genes

rna_seq_data <- rna_seq_data[rna_seq_data[, 4] < 0.05, ]
up_data <- rna_seq_data[rna_seq_data[, 2] > 0, ]
down_data <- rna_seq_data[rna_seq_data[, 2] < 0, ]

Divide initial gene set in to intervals. div_genes_to_quantiles function returns list containing resulting intervals and all possible combinations of intervals

up_genes <- div_genes_to_quantiles(up_data, 6)
down_genes <- div_genes_to_quantiles(down_data, 6)

Preprocess annotation data

go_path <- system.file("extdata", "goslim_plant.obo", package = "fsgor")
gaf_path <- system.file("extdata", "gene_association.tair", package = "fsgor")
annot_obj <- prepare_annotaion(onto_path = go_path, annot_path = gaf_path)

Annotate resulting gene sets using par_multiple_sets_annot function

up_annot <- par_multiple_sets_annot(up_genes, annot_obj, threads = 6, gene_amount_border =  10, background_genes = bg_genes, p_adjust_method =  "BY")
down_annot <- par_multiple_sets_annot(down_genes, annot_obj, threads = 6, gene_amount_border =  10, background_genes = bg_genes, p_adjust_method =  "BY")

Identify fold-sepcific GO terms using recognize_fs_terms function

fs_up <- recognize_fs_terms(up_annot, wholeintname = "1-6", listofcolnames = colnameslist, p_adjust_method = "BY", fdr2step = 1, fdr3step = 0.05)
fs_down <- recognize_fs_terms(down_annot, wholeintname = "1-6", listofcolnames = colnameslist, p_adjust_method = "BY", fdr2step = 1, fdr3step = 0.05)

Visualize th results in two ways, using fold_spec_chart and fold_spec_graph functions

fs_chart <- fold_spec_chart(fs_up, fs_down,scope = NULL,  six_perc_labs)

fs_chart

up_graph <- fold_spec_graph(fs_up, annot_obj, reg = "up", scope = "molecular_function")

plot(up_graph, asp = 0)

References

Omelyanchuk, N. A., Wiebe, D. S., Novikova D. D., Levitsky, V. G., Klimova, N., Gorelova, V., C. Weinholdt,G. V. Vasiliev, E. V. Zemlyanskaya, N. A. Kolchanov, A. V. Kochetov, I. Grosse,V. V. Mironova. Auxin regulates functional gene groups in a fold-change-specific manner in Arabidopsis thaliana roots. Scientific Reports, 2017, 7 (2489), doi:10.1038/s41598-017-02476-8

DanWiebe/fsgor documentation built on May 12, 2019, 12:30 a.m.