ashenkin

na.omit.somecols <- function(data, noNAsInTheseCols, allOutputCols = names(data)) {
    # usage: na.omit.somecols(my_dataframe, c("col1", "col2")).  You can also supply a vector of names (allOutputCols) if you just want certain columns returned.
    completeVec <- complete.cases(data[, noNAsInTheseCols])
    return(data[completeVec, allOutputCols])
}

ashenkin

query_higher_taxa_classes <- function(species_list, known = "genus", order = c("dataframe", "unique_sp")) {
    # Pass in a character vector of species, genera, families, or whatever (the search is flexible)
    # Returns a dataframe with the columns: query, db, family, order, subdivision
    # The dataframe returned is guaranteed to be in the same order as the species list passed in if order is "dataframe"
    order = match.arg(order)
    
    library(taxize)
    library(plyr)
    species_list = sub("^([^ ]*).*$","\\1",species_list) # just take the top level name before the space
    # remove short names that clog the taxon query - replace later

ashenkin

# Thanks to Forrest Stevens.  Some of the code here borrowed from him here: https://github.com/ForrestStevens/Scratch/blob/master/swath_slices.R

library(raster)
library(rgdal)
library(sp)
library(rgeos)
library(gtools)
library(ggplot2)
library(plyr)
library(zoo)

ashenkin

# We often fit LMM/GLMM's with scaled variables.  However, making predictions using those models isn't straightforward (at least to me!)
# It turns out that you have to re-scale your prediction data using the same parameters used to scale your original data frame used to fit the model
# See below, and pay special attention to the section where the new data are rescaled.

library(lme4)
library(VGAM)

reps = 3000

dbh = rexp(reps); dbh = dbh/max(dbh) * 100

ashenkin

contr.sum.keepnames <- function(...) {
    # make deviation contrasts that don't lose the names of the factors in the model results
    # from https://stackoverflow.com/questions/10808853/why-does-changing-contrast-type-change-row-labels-in-r-lm-summary
    conS <- contr.sum(...)
    colnames(conS) = rownames(conS)[-length(rownames(conS))]
    conS
}

set_dev_contrasts <- function(df, colname = "site") {
    # Set contrasts to "deviation coding" so site effects are as compared to overall mean across all sites.  I.e., sites together should have a mean 0 effect.

ashenkin

unscale <- function(scaled, scale, center) {    
  # provide either scale & center, or a scaled vector with the proper attributes    
  if (missing(scale) | missing(center)) {        
    stopifnot( c("scaled:center", "scaled:scale") %in% names(attributes(scaled)) )        
    scale = attr(scaled, "scaled:scale")        
    center = attr(scaled, "scaled:center")        
    attr(scaled, "scaled:scale") <- NULL        
    attr(scaled, "scaled:center") <- NULL    
  }    
  unscaled = scaled * scale + center    

ashenkin

# Alexander Shenkin 2023.
# License: CC BY 4.0: https://creativecommons.org/licenses/by/4.0/.  TLDR; Share and adapt with attribution.
#
# This script reads in a csv file of publications, and produces a list of coauthors
# 1) To get the csv of publications, download a bibtex or other archive of desired pubs from ORCID or another publication search tool.
# Note: don't use Google Scholar!  They limit the number of co-authors exported.
# 2) Import that bibtex or other format archive into Zotero
# 3) Export those imported references in a .csv format.
#
# This does not include institutions of the coauthors, unfortunately.  That still has to be done manually.