,
Yves Van de Peer [aut] | Title: | Motif Analysis In Gene Regulatory Networks |
|---|---|
| Description: | magrene allows the identification and analysis of graph motifs in (duplicated) gene regulatory networks (GRNs), including lambda, V, PPI V, delta, and bifan motifs. GRNs can be tested for motif enrichment by comparing motif frequencies to a null distribution generated from degree-preserving simulated GRNs. Motif frequencies can be analyzed in the context of gene duplications to explore the impact of small-scale and whole-genome duplications on gene regulatory networks. Finally, users can calculate interaction similarity for gene pairs based on the Sorensen-Dice similarity index. |
| Authors: | FabrÃcio Almeida-Silva [aut, cre]
,
Yves Van de Peer [aut] |
| Maintainer: | FabrÃcio Almeida-Silva <[email protected]> |
| License: | GPL-3 |
| Version: | 1.9.0 |
| Built: | 2024-11-18 03:24:35 UTC |
| Source: | https://github.com/bioc/magrene |
Calculate Z-score for motif frequencies
calculate_Z(observed = NULL, nulls = NULL)calculate_Z(observed = NULL, nulls = NULL)
observed |
A list of observed motif frequencies for each motif type. List elements must be named 'lambda', 'bifan', 'V', 'PPI_V', and 'delta' (not necessarily in that order). |
nulls |
A list of null distributions for each motif type
as returned by |
A numeric vector with the Z-score for each motif type.
# Simulating it for test purposes null <- rnorm(1000, mean = 5, sd = 1) nulls <- list( lambda = null, V = null, PPI_V = null, delta = null, bifan = null ) observed <- list(lambda = 7, bifan = 13, delta = 9, V = 5, PPI_V = 10) z <- calculate_Z(observed, nulls) # Check for motif enrichment (Z > 5) z[which(z > 5)]# Simulating it for test purposes null <- rnorm(1000, mean = 5, sd = 1) nulls <- list( lambda = null, V = null, PPI_V = null, delta = null, bifan = null ) observed <- list(lambda = 7, bifan = 13, delta = 9, V = 5, PPI_V = 10) z <- calculate_Z(observed, nulls) # Check for motif enrichment (Z > 5) z[which(z > 5)]
Find bifan motifs
find_bifan( edgelist = NULL, paralogs = NULL, lambda_vec = NULL, count_only = FALSE )find_bifan( edgelist = NULL, paralogs = NULL, lambda_vec = NULL, count_only = FALSE )
edgelist |
A 2-column data frame with regulators in column 1 and targets in column 2. It can be ignored if you give lambda motifs to parameter lambda_vec (recommended). |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. |
lambda_vec |
A character of lambda motifs as returned
by |
count_only |
Logical indicating whether the function should return only motif counts as a numeric scalar. If FALSE, it will return a character vector of motifs. Default: FALSE. |
A character vector with bifan motifs represented in the format regulator1, regulator2->target1, target2.
data(gma_grn) data(gma_paralogs) edgelist <- gma_grn[1:50000, 1:2] paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] paralogs <- rbind( paralogs, data.frame(duplicate1 = "Glyma.01G177200", duplicate2 = "Glyma.08G116700") ) lambda_vec <- find_lambda(edgelist, paralogs) bifan <- find_bifan(paralogs = paralogs, lambda_vec = lambda_vec)data(gma_grn) data(gma_paralogs) edgelist <- gma_grn[1:50000, 1:2] paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] paralogs <- rbind( paralogs, data.frame(duplicate1 = "Glyma.01G177200", duplicate2 = "Glyma.08G116700") ) lambda_vec <- find_lambda(edgelist, paralogs) bifan <- find_bifan(paralogs = paralogs, lambda_vec = lambda_vec)
Find delta motifs
find_delta( edgelist = NULL, paralogs = NULL, edgelist_ppi = NULL, lambda_vec = NULL, count_only = FALSE )find_delta( edgelist = NULL, paralogs = NULL, edgelist_ppi = NULL, lambda_vec = NULL, count_only = FALSE )
edgelist |
A 2-column data frame with regulators in column 1 and targets in column 2. It can be ignored if you give lambda motifs to parameter lambda_vec (recommended). |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. It can be ignored if you give lambda motifs to parameter lambda_vec (recommended). |
edgelist_ppi |
A 2-column data frame with IDs of genes that encode each protein in the interacting pair. |
lambda_vec |
A character of lambda motifs as returned
by |
count_only |
Logical indicating whether the function should return only motif counts as a numeric scalar. If FALSE, it will return a character vector of motifs. Default: FALSE. |
A character vector with lambda motifs represented in the format target1<-regulator->target2.
data(gma_grn) data(gma_paralogs) data(gma_ppi) edgelist <- gma_grn[500:1000, 1:2] # reducing for test purposes edgelist <- gma_grn[1:10000, 1:2] paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] edgelist_ppi <- gma_ppi lambda_vec <- find_lambda(edgelist, paralogs) motifs <- find_delta(edgelist_ppi = edgelist_ppi, lambda_vec = lambda_vec)data(gma_grn) data(gma_paralogs) data(gma_ppi) edgelist <- gma_grn[500:1000, 1:2] # reducing for test purposes edgelist <- gma_grn[1:10000, 1:2] paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] edgelist_ppi <- gma_ppi lambda_vec <- find_lambda(edgelist, paralogs) motifs <- find_delta(edgelist_ppi = edgelist_ppi, lambda_vec = lambda_vec)
Find lambda motifs
find_lambda(edgelist = NULL, paralogs = NULL, count_only = FALSE)find_lambda(edgelist = NULL, paralogs = NULL, count_only = FALSE)
edgelist |
A 2-column data frame with regulators in column 1 and targets in column 2. |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. |
count_only |
Logical indicating whether the function should return only motif counts as a numeric scalar. If FALSE, it will return a character vector of motifs. Default: FALSE. |
A character vector with lambda motifs represented in the format target1<-regulator->target2.
data(gma_grn) data(gma_paralogs) edgelist <- gma_grn[500:1000, 1:2] # reducing for test purposes paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] motifs <- find_lambda(edgelist, paralogs)data(gma_grn) data(gma_paralogs) edgelist <- gma_grn[500:1000, 1:2] # reducing for test purposes paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] motifs <- find_lambda(edgelist, paralogs)
Find V motifs in protein-protein interactions
find_ppi_v(edgelist = NULL, paralogs = NULL, count_only = FALSE)find_ppi_v(edgelist = NULL, paralogs = NULL, count_only = FALSE)
edgelist |
A 2-column data frame with protein 1 in column 1 and protein 2 in column 2. |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. |
count_only |
Logical indicating whether the function should return only motif counts as a numeric scalar. If FALSE, it will return a character vector of motifs. Default: FALSE. |
This function aims to find the number of paralogous gene pairs that share an interaction partner.
A character vector with V motifs represented in the format paralog1-partner-paralog2.
data(gma_ppi) data(gma_paralogs) edgelist <- gma_ppi paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] motifs <- find_ppi_v(edgelist, paralogs)data(gma_ppi) data(gma_paralogs) edgelist <- gma_ppi paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] motifs <- find_ppi_v(edgelist, paralogs)
Find V motifs
find_v(edgelist = NULL, paralogs = NULL, count_only = FALSE)find_v(edgelist = NULL, paralogs = NULL, count_only = FALSE)
edgelist |
A 2-column data frame with regulators in column 1 and targets in column 2. |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. |
count_only |
Logical indicating whether the function should return only motif counts as a numeric scalar. If FALSE, it will return a character vector of motifs. Default: FALSE. |
A character vector with V motifs represented in the format regulator1->target<-regulator2.
data(gma_grn) data(gma_paralogs) edgelist <- gma_grn[2000:4000, 1:2] # reducing for test purposes paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] motifs <- find_v(edgelist, paralogs)data(gma_grn) data(gma_paralogs) edgelist <- gma_grn[2000:4000, 1:2] # reducing for test purposes paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] motifs <- find_v(edgelist, paralogs)
Generate null distributions of motif counts for each motif type
generate_nulls( edgelist = NULL, paralogs = NULL, edgelist_ppi = NULL, n = 1000, bp_param = BiocParallel::SerialParam() )generate_nulls( edgelist = NULL, paralogs = NULL, edgelist_ppi = NULL, n = 1000, bp_param = BiocParallel::SerialParam() )
edgelist |
A 2-column data frame with regulators in column 1 and targets in column 2. |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. |
edgelist_ppi |
A 2-column data frame with IDs of genes that encode each protein in the interacting pair. |
n |
Number of degree-preserving simulated networks to generate. Default: 1000. |
bp_param |
BiocParallel back-end to be used. Default: BiocParallel::SerialParam(). |
A list of numeric vectors named lambda, delta, V,
PPI_V, and bifan, containing the null distribution of motif counts
for each motif type.
set.seed(123) data(gma_grn) data(gma_paralogs) data(gma_ppi) edgelist <- gma_grn[500:1000, 1:2] # reducing for test purposes paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] edgelist_ppi <- gma_ppi n <- 2 # small n for demonstration purposes generate_nulls(edgelist, paralogs, edgelist_ppi, n)set.seed(123) data(gma_grn) data(gma_paralogs) data(gma_ppi) edgelist <- gma_grn[500:1000, 1:2] # reducing for test purposes paralogs <- gma_paralogs[gma_paralogs$type == "WGD", 1:2] edgelist_ppi <- gma_ppi n <- 2 # small n for demonstration purposes generate_nulls(edgelist, paralogs, edgelist_ppi, n)
The GRN was inferred with BioNERO using expression data from Libault et al., 2010, and Severin et al., 2010.
data(gma_grn)data(gma_grn)
A 3-column data frame with node1, node2, and edge weight.
Severin, A. J., Woody, J. L., Bolon, Y. T., Joseph, B., Diers, B. W., Farmer, A. D., ... & Shoemaker, R. C. (2010). RNA-Seq Atlas of Glycine max: a guide to the soybean transcriptome. BMC plant biology, 10(1), 1-16.
Libault, M., Farmer, A., Joshi, T., Takahashi, K., Langley, R. J., Franklin, L. D., ... & Stacey, G. (2010). An integrated transcriptome atlas of the crop model Glycine max, and its use in comparative analyses in plants. The Plant Journal, 63(1), 86-99.
data(gma_grn)data(gma_grn)
The repertoire of soybean paralogs was retrieved from Almeida-Silva et al., 2020.
data(gma_paralogs)data(gma_paralogs)
A 3-column data frame with duplicate 1, duplicate 2, and duplication type
Almeida-Silva, F., Moharana, K. C., Machado, F. B., & Venancio, T. M. (2020). Exploring the complexity of soybean (Glycine max) transcriptional regulation using global gene co-expression networks. Planta, 252(6), 1-12.
data(gma_paralogs)data(gma_paralogs)
PPI were retrieved from the STRING database and filtered to keep only medium confidence edges and nodes in the GRN.
data(gma_ppi)data(gma_ppi)
A 2-column data frame with node1 and node2.
data(gma_ppi)data(gma_ppi)
Data were filtered exactly as demonstrated in the vignette. Briefly, the top 30k edges from the GRN were kept, and only WGD-derived gene pairs were used.
data(nulls)data(nulls)
A list of numeric vectors with the motif frequencies in each simulated network. List elements are named lambda, delta, V, PPI_V, and bifan, and each element has length 100.
data(nulls)data(nulls)
Calculate Sorensen-Dice similarity between paralogous gene pairs
sd_similarity(edgelist = NULL, paralogs = NULL)sd_similarity(edgelist = NULL, paralogs = NULL)
edgelist |
A 2-column data frame with regulators in column 1 and targets in column 2. |
paralogs |
A 2-column data frame with gene IDs for each paralog in the paralog pair. |
A data frame containing the paralogous gene pairs and their Sorensen-Dice similarity scores.
data(gma_ppi) data(gma_paralogs) edgelist <- gma_ppi paralogs <- gma_paralogs sim <- sd_similarity(edgelist, paralogs)data(gma_ppi) data(gma_paralogs) edgelist <- gma_ppi paralogs <- gma_paralogs sim <- sd_similarity(edgelist, paralogs)