The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.
custom.subsets, chrono.subsets, boot.matrix and dispRity. This change is not affecting the user level appart from now allowing to bypass some error messages (thanks to Mario Corio for that obvious suggestion).name.subsets for directly accessing the subsets names of a dispRity object (basically doing names(my_dispRity$subsets)).MCMCglmm.variance for calculating the variance for specific terms in a "MCMCglmm" model.pgls.dispRity to run PGLS test on a dispRity object with a level-2 metric and a tree (using excellent phylolm algorithm). The new test comes with its own S3 print, summary and plot functions if the input dispRity data contains multiple trees or multiple matrices (running and handling the output of multiple phylolm).get.tree utility function to get the trees in each subsets (thanks to Jack Hadfield for this suggestion and help with implementation).roundness to measure the roundness of a variance-covariance matrix.custom.subsets, chrono.subsets, boot.matrix and dispRity functions: the tree, FADLAD, prob and dimensions arguments is now NULL by default (instead of missing).projections function. The defaults are now scale = TRUE, centre = FALSE (previously TRUE) and abs = FALSE (previously TRUE). The default arguments for dispRity.covar.projections remain unchanged though (scale = TRUE, centre = TRUE, abs = TRUE).randtest.dispRity function’s subsets argument now can take a list of arguments of subsets to compare is data is a dispRity object. The call message has also been updated to be much more verbose and clear of what has been tested.projections.* functions for speed.test.metric now also adds the slope coefficient if the option model is used.plot/summary on test.metric results are now more easy to interpret.reduce.space has now 25 times less chances to be stuck on a local optimum.match.tip.edges now also works on nodes and on "multiPhylo" objects and has a added an option use.parsimony (default is TRUE) to propagate the matching down the tree (i.e. if two nodes have the same unique ancestor and variable value, the ancestor is consider to have that variable value as well).covar.plot.clean.data can now also handle a list of "phylo" objects as tree input (i.e. it does not need to be specifically classed as "multiPhylo").spptest to GET.space.maker can now generate specific row names and be replicated to generate a bunch of spaces (via the new optional arguments elements.names and replicates respectively).add.tree utility function now has an optional argument to override any existing trees (replace = TRUE) or not (replace = FALSE; which remains the default previous behaviour).print.dispRity now recognises and highlights simulated data from the treats package.NA management for S3 functions.covar.plot making the figures more aesthetic.remove.zero.brlen now also removes negative branch lengths and works on "multiPhylo" objects (thanks to Thomas Johnson for this suggestion).fill.dispRity has now an extra argument check to toggle the data checking on and off (more for developers).multi.ace is now described in details in the manual.slice.tree has now a keep.all.ancestors option to do exactly that.test.metric.get.disparity now correctly concatenates one dimensional results into a "numeric" (rather than a "matrix").make.metric now internally handles covar object correctly (i.e. as distance matrices).ellipse.volume has been changed to ellipsoid.volume to more accurately reflect what it is measuring.rescale.dispRity has been changed to scale.dispRity and correctly registered as a S3 method.randtest.dist has been changed to distance.randtest to avoid conflict with *.dist S3 methods.select.axes for selecting and analysing the number of axes required to contain an arbitrary amount of variance.randtest.dist for measuring the distance between the observed statistic and a specific quantile of the simulated statistic (thanks to Frane Babarovic for the inspiration).dispRity objects can now contain covariance matrices as a $covar object. The covar part can be directly used for some specific metrics (usually my_metric.covar) and are handled by the dispRity function (and plot, summary, etc…) in a specific way. $covar contains a list of two elements VCV the variance covariance matrix and loc the coordinates of the centre of the VCV in space (can be left empty). $covar matrices are effectively treated as bootstraps.covar.plot for plotting the covar content of dispRity objects (this is separated from plot.dispRity because of the many different options).MCMCglmm.subsets is a function that allows to convert a MCMCglmm object into a dispRity object.projections.between a between group metric for applying the projections metric between the major covariance axis of two matrices.disalignment: the (dis)alignment of a group compared to another one (i.e. the rejection from group B’s centre on the group A’s major axis). a between group metric for applying the projections metric between the major covariance axis of two matrices.dispRity.fast function for the fastest disparity calculations at the expanses of pretty much everything this package does. This is a really situational function.MCMCglmm objects: MCMCglmm.traits for extracting the number of traits, MCMCglmm.levels for extracting the level names, MCMCglmm.sample for sampling posterior IDs and MCMCglmm.covars for extracting variance-covariance matricesdispRity objects with covar matrices: get.covar to extract the VCV matrices (or a subsample of them); axes.covar to extract the major axes of the VCV matrices and as.covar to transform dispRity metric function to use a covar object.match.tip.edge for matching tip labels as factors/characters/integer to tree edges.dispRity.covar.projections for covariance projections analyses (with its associated S3 sub-class).charadriiformes, a data.frame and a phylo object of 359 Charadriiformes species (gulls, plovers and sandpipers) from Cooney et al 2017 along with a MCMCglmm model with each clade as a random term.... in all the dispRity plotting functions can now be targeted to a specific plotting element. When multiple elements are plot by default (e.g. lines, legend, points, etc…) it is now possible to pass a specific ... argument to the specific plotted element using the syntax <element>.<argument> (e.g. points.col = "blue" will only apply the argument col = "blue" to the points).projections and projections.tree metrics: the default "position" output is now scaled, centred and absolute (see ?projections for details).custom.subsets and select.axes. This can create some minor user level changes namely: warning messages for empty subsets now correctly mentions “subsets” (rather than subsamples); groups with incorrect elements are now always flagged as errors (rather than just ignored). The changes at the developer level is that the logic is now made smoother and exported in custom.subsets_fun.R.function.index.csv list (and updater) to help developers find internal functions locations easily.test.dispRity to only occur when using a test of class "htest".custom.subsets can now group elements using a "factor" vector.MCMCglmm objects, dispRity objects in general, dispRity objects with subsets, ect…). It should now be much easier to find these sometimes overlooked functions.dispRity manual, including a section on covar and between.groups specific analyses.test.metric (using the new official dispRity.fast function).data$disparity now don’t have dimension names anymore (significantly reducing the size of disparity objects). However, you can always retrieve the dimensions names using get.disparity.ellipse.volume, you can now directly specify one of the following methods: "pca" to calculate the eigen values from the ordinated matrix; "eigen" to directly do an eigen decomposition of the matrix (new); or "axes" to directly measure the axes (new); or directly provide the eigen values.check.subsets now handles the checking of "dispRity" objects much faster making most functions relying on it slightly faster (this function is typically not called more than once per function).adonis.dispRity to the newest vegan::adonis2 code (thanks to Jari Oksanen for the notification).geiger for dtt.dispRity to avoid package maintenance errors. This leads to no changes at the user level and geiger::dtt is still acknowledged in the manual.tree.age function’s manual now makes it clear it does not estimate tree ages.plot.dispRity(..., type = "preview"), group’s colour attribution and plotting is now made so that the groups larger groups are plotted in the background and the smaller in the foreground.NAs are now better handled in internal checking functions.dispRity when selecting a specific number of dimensions (old warning artefact).plot.dispRity when using type = "preview" on bootstrapped data and for type = "box" when the subsets to plot are from different sizes (now plots all the data correctly).chrono.subsets with "continuous" method a FADLAD data containing only node values (now correctly taken into account; thanks to Peng-Wei Li for noticing it) and when using chrono.subsets with "gradual.*" models on empty subsets.standardGeneric functions are now correctly interpreted as functions throughout the package.observed = TRUE is now used as the default).test.metric plots with save.steps options with more than two types of shifts.null.test which is now correctly managing the number of dimensions inherited from dispRity objects (thanks to Alex Slavenko for spotting this one and the two above).is(data, c("array", "matrix")) to is.array(data) for R 4.1.2.matrix.dispRity and extract.dispRity are now deprecated. You should now use respectively get.matrix and get.disparity instead (the arguments don’t change though).projections that allows to measure elements’ projection on an arbitrary axis (or their distance from this axis with measure = "distance").projections.tree that allows to measure elements’ projection on axis between elements of a given tree.edge.length.tree the edge length from each element given a tree (with the option to.root = TRUE/FALSE to measure the edge length from the element to the root of the tree (default = TRUE) or the nearest ancestor (FALSE).test.metric with save.steps and then visualise them with plot.dispRity along side the disparity metric test results.n.subsets to directly get the number of subsets in a dispRity object.randtest.dispRity that is a wrapper for ade4::randtest applied to dispRity objects (not dissimilar from null.test).dispRity objects now have a reserved $tree component that contain any number of trees attached to the data. This allows any function to use the reserved argument name tree to extract directly the relevant tree from the dispRity object, for functions like chrono.subsets or metrics like ancestral.dist! To help manipulate the tree component of the dispRity object, you can now use the new utility functions add.tree, get.tree and remove.tree.4.0 back to 3.6 following Joseph Brown’s issue and fix.reduce.space "random" algorithm now outputs a named logical vector (like the other algorithms!)."only" text when printing dispRity objects that contains “only” matrices (even though that can be 10k matrices!).summary.dispRity for "dispRity" "randtest" objects to output “ready-to-publish” result tables.estimation.details argument to multi.ace allowing to also return specific arguments from the ancestral states estimation (thanks to Armin Elsler for the suggestion).inc.nodes to clean.data whether to check if the nodes in the tree match the labels in the matrix.make.metric with the option silent = TRUE now outputs a list of info rather than only the level of the metric. You can reproduce the old behaviour using make.metric(..., silent = TRUE)$type).plot using preview when the given argument pch did not match the number of groups (the different pch arguments are now used correctly).ancestral.dist metric. The function is now much faster and much easier to use (due to the new dispRity object structure). The options nodes.coords has been removed and the option full is now changed by to.root. If you still really want to use the older version of ancestral.dist using ancestral.dist.deprecated though.dimensions option throughout the package (e.g. in the dispRity function) can now also be a vector of dimensions to take into consideration (e.g. c(1,2,5)).chrono.subsets now automatically detects the number of digits to round for the internal time slicing functions (thanks to Mario Corio for finding this one).test.metric plots that now display correctly the “top” and “bottom” changes for the “position” shift.test.metric plots that now display the R^2 values correctly.tree.age when the tree tips/node labels vector is longer than the actual number of tips/nodes in the tree.ancestral.dist (see NEW FEATURES above).node.coordinates function (no replacement; you must use a package version prior 1.5.10 to use this function).get.ancestors function (no replacement; you must use a package version prior 1.5.10 to use this function).multi.ace for performing fast ancestral character estimations on multiple matrices (based on castor::asr_mk_model).reduce.space, a function to modify trait spaces imported from the moms shiny app. This function comes with a new reduction algorithm: the “evenness” algorithm for flattening the curve (thanks to Gavin Thomas for the suggestion).test.metric (and associated plot, print and summary functions), to apply the reduce.space function on a specific space and metric to test whether a metric is picking up specific changes in trait space.dispRity function can now use "between.groups" metrics to calculate disparity between groups rather than within groups. The make.metric function is now modified to allow detection of metrics that can be applied between groups.group.dist, a dimension level 1 metric for between groups that measures the distance between two groups. By default, this is the minimum distance but the function takes the probs argument allowing the distance to be between, says, the 95% CI (probs = c(0.025, 0.975))) or between the centroids (probs = c(0.5)).point.dist, a dimension level 2 metric for between groups that measures the distance between the rows in matrix to a point in matrix2. That point is the centroid by default but the point argument can take any function.dispRity package now depends on R (>= 4.0.0).dispRity manual.plot.dispRity S3 methods (for a potential ggpRity?). These changes should not be apparent at the user level but see the two removed options below:plot.dispRity: the chrono.subsets option (TRUE/FALSE) has now been removed. The time-slicing-ness is now automatically detected or can be specified by the user normally through xlab.plot.dispRity: the ylim, ylab, xlab and col options have now been removed. They are now handled through ... as normal generic plot(...) arguments.plot.dispRity: the elements.pch option has now been removed. The pch of the plotted elements can now be passed like other options directly to elements (e.g. elements = list(pch = 15)).plot.dispRity: the dimensions, matrix, nclass and coeff options have now been removed. Any options for dual class plots (randtest, dtt, model.test, type = "preview" etc…) are now handled through the generic specific.args argument.dtt.dispRity: scale.time allowing to scale the time axis (like in geiger::dtt) or not.chrono.subsets dispRity objects, the x label ticks are now rounded if possible (for nicer looking plots!).chrono.subsets time slices, the name of the time slices (their age) is now rounded for aesthetics.Claddis package to match version 0.6 (thanks to Graeme Lloyd for fixing this one!)Claddis.ordination function to be compatible with the new ape version 5.4 (thanks to Emmanuel Paradis for the pointing that out).chrono.subsets where ladder trees with no tip branch lengths an method = "continuous" option would get stuck in an infinite loop.chrono.subsets where the "*.split" methods would bug if the last slice is through a single edge.dispRity where some 1D matrices could loose their class for certain metrics.angles, a dimension level 2 metric that measures the angle of the main axis of each dimension in a matrix (in slopes, angles or degrees).deviations, a dimension level 2 metric that measures the deviation of each element from a hyperplane.char.diff function. It now uses a way faster bitwise comparison architecture and comes with different distance methods as well as modular optional arguments on how to treat various special tokens ("?", "-", "&", etc.). This also allows many more distance methods (now including "hamming", "manhattan", "comparable", "euclidean", "maximum" and "mord").dispRity functions can now intake a single "matrix" or a "list" of matrices with the same row names and dimensions. The disparity is then calculated directly on all the matrices and summarised as before through summary.dispRity. This option can be used to add uncertainty to disparity calculations. For example in chrono.subsets you can now provide a list of trees and a list of associated ancestral state estimates; or for custom.subsets you can provide a list of matrices with different values representing different estimations of the traits.reduce.matrix to work with vegan::vegdist version 2.5-6 (thanks to Jari Oksanen for the fix).R version 4.0.0: class(.) == * is now is(., *).... argument bug PR#16223.make.metric the argument ... is now ignored if any names(...) is "tree" or "phy".neighbours and span.tree.length when feeding “distance” like metrics (thanks to Ashley Reaney for finding that one).chrono.subsets with method = "continuous (now > 1000 times faster!).plot.dispRity with time slices.paleotree package dependency (though the links to this excellent package are still there!).R version requirement to 3.5.... in summary.dispRity are now directly passed to cent.tend (e.g. na.rm = TRUE).castor package.dispRity object architecture (see above). Note that this might effect the exact results of calculations using these demo datasets.make.metric through the data.dim option.dispRity are now tested using the input data dimensions.chrono.subsets with multiple trees now stretches the root edges length to match the oldest tree.displacements, a dimension level 2 metric that measures the position of elements in space (i.e. their distance from the centre relative to their distance to their centroid).neighbours, a dimension level 2 metric that measures the distance from an element to its neighbour (e.g. the nearest neighbour, the furthest, the median, etc.).quantiles, a dimension level 2 metric that measures the nth quantile range per axis (a good alternative to the ranges function!).func.eve, a dimension level 1 metric that measures the functional evenness (i.e. the spread along the minimum spanning tree; from Villéger et al. 2008).func.div, a dimension level 1 metric that measures the functional divergence (i.e. the ratio of deviation from the centroid; from Villéger et al. 2008).span.tree.length now outputs the length of each edges (c.f. the sum of the length) and becomes a level 2 metric.chrono.subsets can now take multiPhylo objects for slicing through multiple trees at once!reduce.matrix for optimising data overlap in a matrix with missing data.slide.nodes for sliding specific nodes on a tree.remove.zero.brlen for stochastically removing zero branch lengths on a tree (using the slide.nodes function).plot.dispRity: the type argument can now be "preview" to have a glimpse at two of the dimensions of the trait-space.Claddis.ordination can now directly take a matrix’s path as input (leaving the function to read and transform the matrix into Claddis format. The function can thus now also be used to convert matrices into Claddis format.centroids and ancestral.dist functions can now take the method option for "euclidean" or "manhattan" distances.switch.dispRity are more verbose when input the wrong metric(s).scree option in space.maker does not require to sum up to one anymore.cor.matrix option in space.maker does not require to have a valid Choleski decomposition (an approximation is used instead).clean.data that did not output dropped tips correctly when applied on multiPhylo objects.chrono.subsets for funky time slices/bins (e.g. with negative values).time.slice function.NA as na.rm or na.omit.dispRity objects now contain a metric argument (if a metric was applied). This argument can now be recycled by the appropriate functions (e.g. in null.test).plot.dispRity argument observed can now take a list of arguments to be passed points().boot.matrix now makes a error warning message when bootstrapping distance matrices (as suggested by Dave Bapst!).geomorph.ordination can now be used to simply create coordinates matrices (no ordination!) with ordinate = FALSE argument.faster eigen value estimations in ellipse.volume when the argument is left missing.Claddis.ordination function. This function now uses the brand new version of the Claddis package on CRAN (0.3).plot.dispRity with option "box" now correctly display plot ranges when disparity is an observed distribution.test.dispRity handles errors messages more efficiently when disparity is an observed distribution.summary.dispRity handles non-bootstrapped distributions display properly.geomorph.ordination now converts "character" vectors into "factors".adonis.dispRity now properly handles complex formulas (with arithmetic signs).... are now properly handled by internal metric testing functions for more accurate error messages.char.diff names are now properly protected in the C implementation to comply with new rcheck requirements.model.test, model.test.sim and model.test.wrapper for fitting models of disparity evolution through time (with associated manuals, vignettes and S3 methods! Thanks to Mark Puttick).boot.matrix: prob for passing probabilities of sampling for specific elements.print method for objects of class "dtt" and "dispRity" (from dtt.dispRity).dtt.dispRity now allows to specify the alternative hypothesis (if nsim > 0).ellipse.volume can now take an explicit eigen value vector (the eigen values are still automatically estimated correctly for PCO and MDS).make.metric when dealing with optional arguments.plot.dispRity.dtt.char.diff to properly reflect the probability of different splits between characters (thanks to Abigail Pastore).Claddis.ordination function.Claddis.ordination function (with add = TRUE).C symbols properly.S3 methods.scale.dispRity is now rescale.dispRity.merge.subsets is now combine.subsets.time.subsets is now chrono.subsets - time.subsets can still be called as an alias for the same function.Claddis.ordination function to comply with the CRAN requirement (this function is still live in the GitHub version 1.1).custom.subset can now automatically create clade-based groups if a phylo object is passed to group.extinction.subsets, to get the list to be passed to test.dispRity for testing the effect of extinction.dtt.dispRity, a wrapper for geiger::dtt. This version is slower that geiger::dtt but allows any univariate disparity metric!adonis.dispRity, a wrapper for vegan::adonis.crown.stem for separating a tree into crown and stem groups.span.tree.length the length of the minimum spanning tree.pairwise.dist: the element’s pairwise distances.radius: the radius of each dimensions.n.ball.volume: the n-dimensional sphere or ellipsoid volume.time.subsets, model = "equal.split" and model = "gradual.split" that retain the probability of being either the descendant or the ancestor. This probability is passed to boot.matrix .stats::dist to vegan::vegdist to allow more distances to be passed through methods arguments.slice.tree can now slice through a single edge.tree.age to estimate ages for trees with fossils only.subsample is now replaced by subset (e.g. time.subsamples is now renamed time.subsets, data$subsamples is now data$subsets, etc…).time.subsets, model = "gradual" is now replaced by model = "proximity" and model = "punctuated" is now replaced by model = "random".ancestral.distance to get the distance from taxa/nodes to their ancestors.random.circle for generating random circle coordinates (see example in space.maker for creating doughnut spaces!).get.bin.ages for getting the geological timescale of a tree (based on geoscale).t0 argument to time.subsamples allowing to set the start age of the first subsample.slice.tree.Claddis.ordination and geomorph.ordination for automatically ordinating data from Claddis and geomorph packages!char.diff for calculating character differences and associated plot function (plot.char.diff)merge.subsamples for… merging subsamples.size.subsamples for getting the size of subsamples in a disparity object.dispRity.through.time and dispRity.per.group now runs easy default disparity analysis.n*(n-1). Bigger matrices now only trigger a warning.dimensions optional argument to dispRity to overwrite the number of dimensions generated by boot.matrix.variances, ranges and centroids are now simplified for speed. The optional arguments for data cleaning are now passed to make.metric.space.maker now allows to approximate the dimensions variance distribution with the scree option.hyper.volume metric for dependencies reasons,parallel option from boot.matrix (the new architecture is already super fast: <2sec for 5k taxa and 10k bootstraps!).series as a part of dispRity objects is now changed to subsamples throughout the whole package.time.series is now renamed time.subsamples, if dates are provided and method is discrete, this function doesn’t need a phylogeny any more.get.subsamples.dispRity is now renamed get.subsamples.cust.series is now renamed custom.subsamples (to avoid confusion with custard.subsamples!). Its factor argument as been changed to groups and can now take a simple list.dispRity object architecture.sim.morpho can now use model = "mixed" for using both HKY-binary and Mk in characters simulation.dispRity object utilities are now all grouped under the ?dispRity.utilities manual with appropriate S3 methods.rm.last.axis argument in boot.matrix. It is now replaced by dimensions.plot.dispRity, type = "lines" is now replaced by type = "line".merge.time.series for cleaning or merging time series,.dispRity-simulate_data on how to simulate morphological characters in dispRity.sim.morpho generates morphological matrices.check.morpho for checking how “realistic” the simulate morphological matrices are.get.contrast.matrix and apply.inapplicable functions for morphological matrices.pair.plot, scale.dispRity and sort.dispRity.space.maker for creating some multidimensional spaces!convhull.surface, convhull.volume and hyper.volume.null.test.plot.dispRity arguments: density for controlling the polygons density and add for adding plots.dispRity, test.dispRity, plot.dispRity, and summary.dispRity can now intake one or more distribution rather than just one or more single values of disparity; whether the data is bootstrapped or not).dispRity can now intake dispRity objects with level 2 disparity metrics.boot.matrix and dispRity can now run in parallel.centroids disparity metric can now use a centroid argument for fixing the centroid point value.variances and ranges disparity metrics can now intake a k.root argument for scaling the results.type_discrete argument in plot.dispRity and type argument can now be:
continuous disparity curves.box for real boxplots.lines for the distribution vertical lines.polygon for the distribution boxes.get.dispRity for subsampling dispRity objects.extract.dispRity for extracting disparity results.test.dispRity for applying tests to dispRity objects.make.metric for helping creating your very own disparity metric.hyper.volume for measuring the morphospace hyper-ellipsoid volume.metric argument from dispRity can now intake up two three functions (see dispRity.metric and make.metric).plot.dispRity options improved (rarefaction + default).cust.series can now intake multiple factors columns.boot.matrix, dispRity, summary and plot now also include observed values.plot now has an observed option to plot the observed disparity.taxa to elements.plot option diversity has been renamed elements.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.