Changes in version 1.3.3 - July 19, 2024 - fix in .read_matrix_block() - See https://github.com/GabrielHoffman/dreamlet/pull/23/commits/c453ac98ebc0329279b4dd3ae26a674df0e9b1f2 Changes in version 1.3.1 - May 31, 2024 - bump Bioc version Changes in version 1.1.24 - May 28, 2024 - dreamlet() gives clearer error message for singular design matrix - plotGeneHeatmap() handles zmax correctly now Changes in version 1.1.23 - May 16, 2024 - processOneAssay, set rescaleWeightsAfter=FALSE by default Changes in version 1.1.22 - May 14, 2024 - fix issue in topTable() with multiple coef values Changes in version 1.1.21 - fix issue in run_mash() with multiple coefficients - from https://github.com/GabrielHoffman/dreamlet/issues/18 Changes in version 1.1.20 - in dreamlet() allow formula to include only intercept Changes in version 1.1.19 - bug fix in compositePosteriorTest() Changes in version 1.1.18 - bug fix Changes in version 1.1.17 - new functionality in plotPCA() and outlierByAssay() - works on any list, not just dreamletProcessedData - allows outlier analysis on residuals Changes in version 1.1.16 - Feb 8, 2024 - fix bug in pbWeights() - Fix scaling issue in outlierByAssay() Changes in version 1.1.15 - Feb 5, 2024 - Fix bug in call to eBayes() - in processAssays() pass argument scaledByLib to voomWithDreamWeights() Changes in version 1.1.14 - Jan 29, 2024 - fix bug in pbWeights() - smaller pseudo variance - limit to only expressed genes by adding getExprGeneNames() Changes in version 1.1.13 - Jan 25, 2024 - improve error reporting in seeErrors() and documentation - update outlier() to compute z-scores. How returns data.frame() - add outlierByAssay() and plotPCA() Changes in version 1.1.12 - Jan 16, 2024 - compositePosteriorTest() allows exclude set to be NULL Changes in version 1.1.11 - Jan 16, 2024 - add meta_analysis() Changes in version 1.1.10 - Jan 10, 2024 - stackAssays() now includeds metadata()$aggr_means correctly - add compositePosteriorTest() Changes in version 1.1.9 - Jan 3, 2024 - use get_metadata_aggr_means() to extract aggr_means when SCE is produced by cbind'ing Changes in version 1.1.8 - Dec 18, 2023 - fix issue when no genes pass cutoffs - fix issue with aggr_means in aggregateToPseudoBulk() - fix bug when rdf is low for all genes Changes in version 1.1.7 - Dec 10, 2023 - add plotBeeswarm() - add rowWeightedVarsMatrix() - bug fixes - add isFullRank() check in dreamlet() - handle exceptions in run_mash() Changes in version 1.1.6 - Dec 5, 2023 - pbWeights() add argument maxRatio Changes in version 1.1.5 - Nov 28, 2023 - fix edge cases in pbWeights() - cell weights is not default in dreamlet() Changes in version 1.1.4 - Nov 27, 2023 - add pbWeights() to compute precision weights for pseudobulk counts - extend extractData() and include it in vignette Changes in version 1.1.3 - Nov 22, 2023 - add stackAssays() - add diffVar() - fix getVarFromCounts) so zeta is a mean, not a sum Changes in version 1.1.2 - Nov 13, 2023 - computeLogCPM() uses augmentPriorCount() Changes in version 1.1.1 - bug fix in Bioc 3.18 and devel Changes in version 0.99.33 - computeLogCPM() now returns matrix instead of sparseMatrix Changes in version 0.99.32 - Oct 18, 2023 - update variancePartition version dependency - add getWeightsList() Changes in version 0.99.31 - Oct 12, 2023 - fix error reporting in processAssays() Changes in version 0.99.30 - Sept 22, 2023 - Fix issue with precision weights from cell number - Include setAutoBlockSize() update within aggregateToPseudoBulk() Changes in version 0.99.28 - Sept 5, 2023 - Update error handling for processAssays() and fitVarPart() Changes in version 0.99.26 - August 18, 2023 - Update error handling and documentation Changes in version 0.99.23 - August 8, 2023 - run styler::style_pkg() Changes in version 0.99.22 - August 8, 2023 - dreamletCompareClusters() now allows cell-level covariates in response to https://github.com/GabrielHoffman/dreamlet/issues/11 - Fix code for Bioconductor submission Changes in version 0.99.21 - July 17, 2023 - Improve functionality and documentation of dreamlet::residuals() Changes in version 0.99.20 - June 29, 2023 - in processAssays() use voomWithDreamWeights(..., span="auto") to estimate the lowess tuning parameter - rare error in merge_metadata() when a cell type is not observed for all donors. Changes in version 0.99.19 - June 28, 2023 - in dreamlet() fix issue when contrasts are specified and formula includes variable from metadata() Changes in version 0.99.18 - June 27, 2023 - add assays argument to buildClusterTreeFromPB() Changes in version 0.99.14 - June 16, 2023 - bug fix in processAssays() when assays is dropped Changes in version 0.99.13 - May 31, 2023 - improved error reporting - Compatibility with variancePartition v2.0.5 (renamed 1.31.1) Changes in version 0.99.12 - May 24, 2023 - required zellkonverter (>= 1.10.1) to avoid issues with previous version - issue solved by https://github.com/theislab/zellkonverter/blob/b56718d113327020c024e188d9ac67ea57eaf35d/R/AnnData2SCE.R#L351 Changes in version 0.99.11 - May 12, 2023 - Compatibility with variancePartition v2.0.1 Changes in version 0.99.10 - April 24, 2023 - Fix issue in raised in Bioconductor submission: https://github.com/Bioconductor/Contributions/issues/2955#issuecomment-1498070980 - Compatibility with variancePartition v2.0.0 Changes in version 0.99.6 - March 29, 2023 - fix topTable() for dreamletResult in the case where one or more cells didn't estimate the coefficient of interest Changes in version 0.99.3 - March 23, 2023 - reduce compiler time - add computeNormCounts() and computeLogCPM() Changes in version 0.99.1 - March 20, 2023 - fix Biocondcutor submission based on https://github.com/Bioconductor/Contributions/issues/2955#issuecomment-1476037237 Changes in version 0.99.0 - March 15, 2023 - Biocondcutor submission Changes in version 0.0.64 - March 7, 2023 - fix topTable() to deal with multiple coef as array - add vignette about non-linear effects - Add colData<- for dreamletProcessedData Changes in version 0.0.63 - March 3, 2023 - small bug fixes in topTable() and plotForest() Changes in version 0.0.62 - Feb 27, 2023 - aggregateToPseudoBulk() stores mean of cell-level covariates in metadata(pb)$aggr_means - Use these covariates in processAssays(), dreamlet(), fitVarPart() - extend to aggregateNonCountSignal() Changes in version 0.0.61 - Jan 25, 2023 - add plotProjection() - add outlier() - update plotForest() Changes in version 0.0.60 - Jan 18, 2023 - allow custom ordering of assays in plots Changes in version 0.0.59 - Jan 12, 2023 - update plotVolcano() to allow scales="free_y" - warning when p-values are zero. Changes in version 0.0.58 - Jan 9, 2023 - RewriteaggregateNonCountSignal() to include filters - Depend only on CRAN and BioC packages Changes in version 0.0.57 - Jan 4, 2023 - Include precision weights in aggregateNonCountSignal() Changes in version 0.0.56 - Jan 3, 2023 - Enable processing of non-count data with aggregateNonCountSignal() Changes in version 0.0.55 - Dec 7, 2022 - in plotGeneHeatmap() drop empty genes Changes in version 0.0.54 - Nov 30, 2022 - add buildClusterTreeFromPB() Changes in version 0.0.53 - Nov 18, 2022 - fix bug in topTable() Changes in version 0.0.52 - Nov 18, 2022 - add as.dreamletResult() - update variancePartition dependency and source Changes in version 0.0.51 - Nov 10, 2022 - in processAssays() and processOneAssay(), add argument min.prop indicating the minimum proportion of retained samples with non-zero counts Changes in version 0.0.50 - Nov 2, 2022 - fix bug in zenith_gsa() for few gene sets Changes in version 0.0.49 - Oct 17, 2022 - add computeCellCounts() Changes in version 0.0.48 - Sept 14, 2022 - add transpose argument to plotGeneHeatmap() - and alpha arugment to plotVoom() - update y axis and outlines in plotVarPart() Changes in version 0.0.47 - update filtering of covariates, especially for when many samples are dropped Changes in version 0.0.46 - add totalCPM column to output of cellTypeSpecificity() to use for filtering. Functions dreamlet::plotHeatmap() plotViolin() and plotPercentBars() now ignore this column Changes in version 0.0.45 - update documentation Changes in version 0.0.44 - add plotGeneHeatmap() - add argument assays to plotVarPart() - add extractData() Changes in version 0.0.43 - faster aggregateToPseudoBulk() by speeding up check in .check_arg_assay() - more flexibility for tabToMatrix() Changes in version 0.0.42 - fix misc issues with plotting and order of facets - fix issue with redundant variables in small sample sizes Changes in version 0.0.41 - fix issue with topTable() when all random effects are dropped Changes in version 0.0.40 - Compatibility with zenith package submitted to Bioconductor. Changes in version 0.0.39 - fix issue with aggregateToPseudoBulk() when summarizing for just 1 sample Changes in version 0.0.38 - add getTreat() for dreamlet() result Changes in version 0.0.37 - droplevels for colData in processAssays() Changes in version 0.0.36 - fixes in processAssays() to detect issues with SCE Changes in version 0.0.35 - remove old code and dependencies - change website theme Changes in version 0.0.34 - bug fix for aggregateToPseudoBulk() with sample ordering Changes in version 0.0.33 - bug fix for dreamletCompareClusters() Changes in version 0.0.32 - add colsum2() using beachmat code. Changes in version 0.0.31 - reduce memory usage in aggregateToPseudoBulk() by fixing `aggregateByColnames() Changes in version 0.0.30 - update run_mash() to combine results across coefs Changes in version 0.0.29 - fix bug in dreamlet::colsum_fast() used in pseudobulk Changes in version 0.0.28 - add da_to_sparseMatrix() Changes in version 0.0.27 - aggregateToPseudoBulk() for DelayedArray now uses colsum_fast() - this is faster then the previous version for DelayedArray Changes in version 0.0.26 - update dreamletCompareClusters(): - now compatable with plotZenithResults() - include flag errorsAsWarnings. If TRUE warns and returns NULL. Changes in version 0.0.25 - change return value for dreamletCompareClusters() to be compatible with zenith_gsa() - fix usage of formula in dreamletCompareClusters() Changes in version 0.0.24 - additional speed up aggregateToPseudoBulk() when a Seurat object is used - uses RcppEigen sparse matrix iterator Changes in version 0.0.23 - dramatic speed up aggregateToPseudoBulk() when a Seurat object is used - uses RcppEigen Changes in version 0.0.22 - Speed up aggregateToPseudoBulk() when a Seurat object is used Changes in version 0.0.21 - Add collapse=TRUE to dreamletCompareClusters() Changes in version 0.0.20 - Fix bug in dreamletCompareClusters() Changes in version 0.0.19 - Fix bug in dreamletCompareClusters() Changes in version 0.0.18 - add min.samples to processAssays(), processOneAssay() - add dreamletCompareClusters() and run_mash() - Fix bug in dreamletCompareClusters() - updated mashr dependency Changes in version 0.0.17 - add run_mash() - add zenith_gsa(), plotVolcano(), plotForest() for results - fix bug in cellTypeSpecificity() for genes with zero reads across all cell types - order of arguments in plotForest() and zenith_gsa() changed for consistancy - expand vignettes - bug fix for removeConstantTerms() when excluded variable string (i.e. tissue) is also a substring of other variables (i.e. tissueStatus) Changes in version 0.0.16 - add residuals() for dreamlet() result - add dreamletPairs() - fix bug in removeConstantTerms() with multiple constant terms - improve usability of cellTypeSpecificity() by adding plotPercentBars() and plotViolin() compatability - fix bug in topTable() when coef is not estimated - add argument assays to dreamlet(), fitVarPart(), and processAssays() Changes in version 0.0.15 - processOneAssay() weights by number of cells - require variancePartition >= 1.25.1 to handle weights in voomWithDreamWeights() - fix bug in topTable() - add plotPercentBars() for class vpDF - misc bug fixes - improve documentation Changes in version 0.0.14 - move count ratio code to crumblr package - use applyQualityWeights() Changes in version 0.0.13 - Oct 25, 2021 - add ilr_composition_test.R Changes in version 0.0.12 - Oct 15, 2021 - update print for dreamletResult using coefNames() - small bug fix - fix bugs in regModel() Changes in version 0.0.11 - Oct 6, 2021 - removeConstantTerms() now doesn't drop terms solely because of NA's - this means that other functions can gracefully warn the user about NA's Changes in version 0.0.10 - Oct 5, 2021 - suppress package startup messages in aggregateToPseudoBulk() - bug fix in removeConstantTerms() - Sept 30, 2021 - call to zenith_gsa() adds argument inter.gene.cor and progressbar - fix output to cellTypeCompositionVarPart() and cellTypeCompositionTest() - fix issue with topTable() where FDR was evaluated on only a subset of genes Changes in version 0.0.8 - Sept 28, 2021 - update docs, logos, TODO - update dreamlet() to handle linear contrasts - removeConstantTerms() now drops categorical variables with only a max of one example per category - Cleaner code for cell composition test Changes in version 0.0.7 - Sept 2, 2021 - add cellTypeCompositionTest() - handle random effects Changes in version 0.0.6 - Sept 1, 2021 - enforce package version requirements Changes in version 0.0.5 - August 25, 2021 - handling of pmetadata by processAssays(), fitVarPart(), and dreamlet() - change defaults for bpparam to SerialParam() - Created new files for code - Create object dreamletResult returned by dreamlet() and used by topTable() - more capable dreamletProcessedData object - add details() functions for dreamletResult, dreamletProcessedData and vpDF - More information about when terms are dropped from formulas - type definition to zenith_gsa Changes in version 0.0.4 - add aggregateToPseudoBulk() for faster access to SingleCellExperiment() backed by H5AD Changes in version 0.0.3 - add argument to processAssays() to include extra meta-data - add subseting with assay() for dreamletProcessedData - adapt plotVoom(), plotVolcano(), plotVarPart() to be more flexiable - fitVarPart() returns DataFrame Changes in version 0.0.2 - check failed regression Changes in version 0.0.1 - Initial version