| Title: | Ecological Analysis and Visualization |
|---|---|
| Description: | Support ecological analyses such as ordination and clustering. Contains consistent and easy wrapper functions of 'stat', 'vegan', and 'labdsv' packages, and visualisation functions of ordination and clustering. |
| Authors: | Toshikazu Matsumura [aut, cre] |
| Maintainer: | Toshikazu Matsumura <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 0.2.1.9000 |
| Built: | 2025-01-29 04:54:10 UTC |
| Source: | https://github.com/matutosi/ecan |
Helper function for clustering methods
Helper function for calculate distance
Add group names to hclust labels.
Add colors to dendrogram
cluster(x, c_method, d_method) distance(x, d_method) cls_add_group(cls, df, indiv, group, pad = TRUE) cls_color(cls, df, indiv, group)cluster(x, c_method, d_method) distance(x, d_method) cls_add_group(cls, df, indiv, group, pad = TRUE) cls_color(cls, df, indiv, group)
x |
A community data matrix. rownames: stands. colnames: species. |
c_method |
A string of clustering method. "ward.D", "ward.D2", "single", "complete", "average" (= UPGMA), "mcquitty" (= WPGMA), "median" (= WPGMC), "centroid" (= UPGMC), or "diana". |
d_method |
A string of distance method. "correlation", "manhattan", "euclidean", "canberra", "clark", "bray", "kulczynski", "jaccard", "gower", "altGower", "morisita", "horn", "mountford", "raup", "binomial", "chao", "cao", "mahalanobis", "chisq", "chord", "aitchison", or "robust.aitchison". |
cls |
A result of cluster or dendrogram. |
df |
A data.frame to be added into ord scores |
indiv, group
|
A string to specify individual and group name of column in df. |
pad |
A logical to specify padding strings. |
cluster() returns result of clustering. $clustering_method: c_method $distance_method: d_method
distance() returns distance matrix.
Inputing cls return a color vector,
inputing dend return a dend with color.
library(dplyr) df <- tibble::tibble( stand = paste0("ST_", c("A", "A", "A", "B", "B", "C", "C", "C", "C")), species = paste0("sp_", c("a", "e", "d", "e", "b", "e", "d", "b", "a")), abundance = c(3, 3, 1, 9, 5, 4, 3, 3, 1)) cls <- df2table(df) %>% cluster(c_method = "average", d_method = "bray") library(ggdendro) # show standard cluster ggdendro::ggdendrogram(cls) # show cluster with group data(dune, package = "vegan") data(dune.env, package = "vegan") cls <- cluster(dune, c_method = "average", d_method = "bray") df <- tibble::rownames_to_column(dune.env, "stand") cls <- cls_add_group(cls, df, indiv = "stand", group = "Use") ggdendro::ggdendrogram(cls)library(dplyr) df <- tibble::tibble( stand = paste0("ST_", c("A", "A", "A", "B", "B", "C", "C", "C", "C")), species = paste0("sp_", c("a", "e", "d", "e", "b", "e", "d", "b", "a")), abundance = c(3, 3, 1, 9, 5, 4, 3, 3, 1)) cls <- df2table(df) %>% cluster(c_method = "average", d_method = "bray") library(ggdendro) # show standard cluster ggdendro::ggdendrogram(cls) # show cluster with group data(dune, package = "vegan") data(dune.env, package = "vegan") cls <- cluster(dune, c_method = "average", d_method = "bray") df <- tibble::rownames_to_column(dune.env, "stand") cls <- cls_add_group(cls, df, indiv = "stand", group = "Use") ggdendro::ggdendrogram(cls)
Calculating diversity
d(x) h(x, base = exp(1))d(x) h(x, base = exp(1))
x, base
|
A numeric vector. |
A numeric vector.
Convert data.frame and table to each other.
df2table(df, st = "stand", sp = "species", ab = "abundance") table2df(tbl, st = "stand", sp = "species", ab = "abundance") dist2df(dist)df2table(df, st = "stand", sp = "species", ab = "abundance") table2df(tbl, st = "stand", sp = "species", ab = "abundance") dist2df(dist)
df |
A data.frame. |
st, sp, ab
|
A string. |
tbl |
A table. community matrix. rownames: stands. colnames: species. |
dist |
A distance table. |
df2table() return table, table2df() return data.frame, dist2df() return data.frame.
tibble::tibble( st = paste0("st_", rep(1:2, times = 2)), sp = paste0("sp_", rep(1:2, each = 2)), ab = runif(4)) %>% dplyr::bind_rows(., .) %>% print() %>% df2table("st", "sp", "ab")tibble::tibble( st = paste0("st_", rep(1:2, times = 2)), sp = paste0("sp_", rep(1:2, each = 2)), ab = runif(4)) %>% dplyr::bind_rows(., .) %>% print() %>% df2table("st", "sp", "ab")
Draw layer construction plot
Add mid point and bin width of layer heights.
Compute mid point of layer heights.
Compute bin width of layer heights.
draw_layer_construction( df, stand = "stand", height = "height", cover = "cover", group = "", ... ) add_mid_p_bin_w(df, height = "height") mid_point(x) bin_width(x)draw_layer_construction( df, stand = "stand", height = "height", cover = "cover", group = "", ... ) add_mid_p_bin_w(df, height = "height") mid_point(x) bin_width(x)
df |
A dataframe including columns: stand, layer height and cover. Optional column: stand group. |
stand, height, cover, group
|
A string to specify stand, height, cover, group column. |
... |
Extra arguments for geom_bar(). |
x |
A numeric vector. |
draw_layer_construction() returns gg object, add_mid_p_bin_w() returns dataframe including mid_point and bin_width columns. mid_point() and bin_width() return a numeric vector.
library(dplyr) n <- 10 height_max <- 20 ly_list <- c("B", "S", "K") st_list <- LETTERS[1] sp_list <- letters[1:9] st_group <- NULL sp_group <- rep(letters[24:26], 3) cover_list <- 2^(0:4) df <- gen_example(n = n, use_layer = TRUE, height_max = height_max, ly_list = ly_list, st_list = st_list, sp_list = sp_list, st_group = st_group, sp_group = sp_group, cover_list = cover_list) # select stand and summarise by sp_group df %>% dplyr::group_by(height, sp_group) %>% dplyr::summarise(cover = sum(cover), .groups = "drop") %>% draw_layer_construction(group = "sp_group", colour = "white")library(dplyr) n <- 10 height_max <- 20 ly_list <- c("B", "S", "K") st_list <- LETTERS[1] sp_list <- letters[1:9] st_group <- NULL sp_group <- rep(letters[24:26], 3) cover_list <- 2^(0:4) df <- gen_example(n = n, use_layer = TRUE, height_max = height_max, ly_list = ly_list, st_list = st_list, sp_list = sp_list, st_group = st_group, sp_group = sp_group, cover_list = cover_list) # select stand and summarise by sp_group df %>% dplyr::group_by(height, sp_group) %>% dplyr::summarise(cover = sum(cover), .groups = "drop") %>% draw_layer_construction(group = "sp_group", colour = "white")
Stand, species, and cover are basic. Layer, height, st_group, are sp_group optional.
gen_example( n = 300, use_layer = TRUE, height_max = 20, ly_list = "", st_list = LETTERS[1:9], sp_list = letters[1:9], st_group = NULL, sp_group = NULL, cover_list = 2^(0:6) )gen_example( n = 300, use_layer = TRUE, height_max = 20, ly_list = "", st_list = LETTERS[1:9], sp_list = letters[1:9], st_group = NULL, sp_group = NULL, cover_list = 2^(0:6) )
n |
A numeric to generate no of occurrences. |
use_layer |
A logical. If FALSE, height_max and ly_list will be omitted. |
height_max |
A numeric. The highest layer of samples. |
ly_list, st_list, sp_list, st_group, sp_group
|
A string vector. st_group and sp_group are optional (default is NULL). Length of st_list and sp_list should be the same as st_group and sp_group, respectively. |
cover_list |
A numeric vector. |
A dataframe with columns: stand, layer, species, cover, st_group and sp_group.
n <- 300 height_max <- 20 ly_list <- c("B1", "B2", "S1", "S2", "K") st_list <- LETTERS[1:9] sp_list <- letters[1:9] st_group <- rep(LETTERS[24:26], 3) sp_group <- rep(letters[24:26], 3) cover_list <- 2^(0:6) gen_example(n = n, use_layer = TRUE, height_max = height_max, ly_list = ly_list, st_list = st_list, sp_list = sp_list, st_group = st_group, sp_group = sp_group, cover_list = cover_list)n <- 300 height_max <- 20 ly_list <- c("B1", "B2", "S1", "S2", "K") st_list <- LETTERS[1:9] sp_list <- letters[1:9] st_group <- rep(LETTERS[24:26], 3) sp_group <- rep(letters[24:26], 3) cover_list <- 2^(0:6) gen_example(n = n, use_layer = TRUE, height_max = height_max, ly_list = ly_list, st_list = st_list, sp_list = sp_list, st_group = st_group, sp_group = sp_group, cover_list = cover_list)
Calculating diversity indices such as species richness (s), Shannon's H' (h), Simpson' D (d), Simpson's inverse D (i).
ind_val( df, stand = NULL, species = NULL, abundance = NULL, group = NULL, row_data = FALSE )ind_val( df, stand = NULL, species = NULL, abundance = NULL, group = NULL, row_data = FALSE )
df |
A data.frame, which has three cols: stand, species, abundance. Community matrix should be converted using table2df(). |
stand, species, abundance
|
A text to specify each column. If NULL, 1st, 2nd, 3rd column will be used. |
group |
A text to specify group column. |
row_data |
A logical. TRUE: return row result data of labdsv::indval(). |
A data.frame.
library(dplyr) library(tibble) data(dune, package = "vegan") data(dune.env, package = "vegan") df <- dune %>% table2df(st = "stand", sp = "species", ab = "cover") %>% dplyr::left_join(tibble::rownames_to_column(dune.env, "stand")) ind_val(df, abundance = "cover", group = "Moisture")library(dplyr) library(tibble) data(dune, package = "vegan") data(dune.env, package = "vegan") df <- dune %>% table2df(st = "stand", sp = "species", ab = "cover") %>% dplyr::left_join(tibble::rownames_to_column(dune.env, "stand")) ind_val(df, abundance = "cover", group = "Moisture")
Check cols one-to-one, or one-to-multi in data.frame
is_one2multi(df, col_1, col_2) is_one2one(df, col_1, col_2) is_multi2multi(df, col_1, col_2) cols_one2multi(df, col, inculde_self = TRUE) select_one2multi(df, col, inculde_self = TRUE) unique_length(df, col_1, col_2)is_one2multi(df, col_1, col_2) is_one2one(df, col_1, col_2) is_multi2multi(df, col_1, col_2) cols_one2multi(df, col, inculde_self = TRUE) select_one2multi(df, col, inculde_self = TRUE) unique_length(df, col_1, col_2)
df |
A data.frame |
col, col_1, col_2
|
A string to specify a colname. |
inculde_self |
A logical. If TRUE, return value including input col. |
is_one2multi(), is_one2one(), is_multi2multi() return a logical. cols_one2multi() returns strings of colnames that has one2multi relation to input col. unique_length() returns a list.
df <- tibble::tibble( x = rep(letters[1:6], each = 1), x_grp = rep(letters[1:3], each = 2), y = rep(LETTERS[1:3], each = 2), y_grp = rep(LETTERS[1:3], each = 2), z = rep(LETTERS[1:3], each = 2), z_grp = rep(LETTERS[1:3], times = 2)) unique_length(df, "x", "x_grp") is_one2one(df, "x", "x_grp") is_one2one(df, "y", "y_grp") is_one2one(df, "z", "z_grp")df <- tibble::tibble( x = rep(letters[1:6], each = 1), x_grp = rep(letters[1:3], each = 2), y = rep(LETTERS[1:3], each = 2), y_grp = rep(LETTERS[1:3], each = 2), z = rep(LETTERS[1:3], each = 2), z_grp = rep(LETTERS[1:3], times = 2)) unique_length(df, "x", "x_grp") is_one2one(df, "x", "x_grp") is_one2one(df, "y", "y_grp") is_one2one(df, "z", "z_grp")
Helper function for ordination methods
ordination(tbl, o_method, d_method = NULL, ...) ord_plot(ord, score = "st_scores", x = 1, y = 2) ord_add_group(ord, score = "st_scores", df, indiv, group) ord_extract_score(ord, score = "st_scores", row_name = NULL)ordination(tbl, o_method, d_method = NULL, ...) ord_plot(ord, score = "st_scores", x = 1, y = 2) ord_add_group(ord, score = "st_scores", df, indiv, group) ord_extract_score(ord, score = "st_scores", row_name = NULL)
tbl |
A community data matrix. rownames: stands. colnames: species. |
o_method |
A string of ordination method. "pca", "ca", "dca", "pcoa", or "nmds". "fspa" is removed, because package dave was archived. |
d_method |
A string of distance method. "correlation", "manhattan", "euclidean", "canberra", "clark", "bray", "kulczynski", "jaccard", "gower", "altGower", "morisita", "horn", "mountford", "raup", "binomial", "chao", "cao", "mahalanobis", "chisq", "chord", "aitchison", or "robust.aitchison". |
... |
Other parameters for PCA. |
ord |
A result of ordination(). |
score |
A string to specify score for plot. "st_scores" means stands and "sp_scores" species. |
x, y
|
A column number for x and y axis. |
df |
A data.frame to be added into ord scores |
indiv, group, row_name
|
A string to specify indiv, group, row_name column in df. |
ordination() returns result of ordination. $st_scores: scores for stand. $sp_scores: scores for species. $eig_val: eigen value for stand. $results_raw: results of original ordination function. $ordination_method: o_method. $distance_method: d_method. ord_plot() returns ggplot2 object. ord_extract_score() extracts stand or species scores from ordination result. ord_add_group() adds group data.frame into ordination scores.
library(ggplot2) library(vegan) data(dune) data(dune.env) df <- table2df(dune) %>% dplyr::left_join(tibble::rownames_to_column(dune.env, "stand")) sp_dammy <- tibble::tibble("species" = colnames(dune), "dammy_1" = stringr::str_sub(colnames(dune), 1, 1), "dammy_6" = stringr::str_sub(colnames(dune), 6, 6)) df <- dplyr::left_join(df, sp_dammy) ord_dca <- ordination(dune, o_method = "dca") ord_pca <- df %>% df2table() %>% ordination(o_method = "pca") ord_dca_st <- ord_extract_score(ord_dca, score = "st_scores") ord_pca_sp <- ord_add_group(ord_pca, score = "sp_scores", df, indiv = "species", group = "dammy_1")library(ggplot2) library(vegan) data(dune) data(dune.env) df <- table2df(dune) %>% dplyr::left_join(tibble::rownames_to_column(dune.env, "stand")) sp_dammy <- tibble::tibble("species" = colnames(dune), "dammy_1" = stringr::str_sub(colnames(dune), 1, 1), "dammy_6" = stringr::str_sub(colnames(dune), 6, 6)) df <- dplyr::left_join(df, sp_dammy) ord_dca <- ordination(dune, o_method = "dca") ord_pca <- df %>% df2table() %>% ordination(o_method = "pca") ord_dca_st <- ord_extract_score(ord_dca, score = "st_scores") ord_pca_sp <- ord_add_group(ord_pca, score = "sp_scores", df, indiv = "species", group = "dammy_1")
Pad a string to the longest width of the strings.
pad2longest(string, side = "right", pad = " ")pad2longest(string, side = "right", pad = " ")
string |
Strings. |
side |
Side on which padding character is added (left, right or both). |
pad |
Single padding character (default is spaces). |
Strings.
x <- c("a", "ab", "abc") pad2longest(x, side = "right", pad = " ")x <- c("a", "ab", "abc") pad2longest(x, side = "right", pad = " ")
BiSS data is formatted as JSON.
read_biss(txt, join = TRUE)read_biss(txt, join = TRUE)
txt |
A JSON string, URL or file. |
join |
A logical. TRUE: join plot and occurrence, FALSE: do not join. |
A data frame.
library(dplyr) # path <- "set file path" path <- "https://raw.githubusercontent.com/matutosi/biodiv/main/man/example.json" read_biss(path)library(dplyr) # path <- "set file path" path <- "https://raw.githubusercontent.com/matutosi/biodiv/main/man/example.json" read_biss(path)
Calculating diversity indices such as species richness (s), Shannon's H' (h), Simpson' D (d), Simpson's inverse D (i).
shdi(df, stand = NULL, species = NULL, abundance = NULL)shdi(df, stand = NULL, species = NULL, abundance = NULL)
df |
A data.frame, which has three cols: stand, species, abundance. Community matrix should be converted using table2df(). |
stand, species, abundance
|
A text to specify each column. If NULL, 1st, 2nd, 3rd column will be used. |
A data.frame. Including species richness (s), Shannon's H' (h), Simpson' D (d), Simpson's inverse D (i).
data(dune, package = "vegan") df <- table2df(dune) shdi(df)data(dune, package = "vegan") df <- table2df(dune) shdi(df)