NEWS.md
add_download_button.EpiArchives and offload report to there:
EpiArchives.plot_enrichment
NULL.workers <- check_workers(workers = workers) to all functions that take workers to handle workers=NULL properly.download_button:
prepare_blacklist:
EpiCompare(blacklist=NULL) is now the default.prepare_genome_builds:
blacklist arg is not a user-supplied GRanges object)mm9_blacklist
width_boxplotplot_enrichmentplot_ChIPseeker_annotationoverlap_stat_plot
annotation arg to more informative txdb arg, and set default to NULL, which ChIPseeker functions will automatically handle.as_interactive:
EpiCompare::EpiCompare arguments:
error: keep knitting even on errors.tss_distance: upstream/downstream of TSS.quiet: knit quietlyEpiCompare command code as text:
width_boxplot:
data.table and lapply
tss_plot:
ChIPseeker::getTagMatrix
methods::show from all partsresults='asis' globally instead of in each chunk header.number_sections: true
{-} tags.plot_chromHMM:
Error in (function (classes, fdef, mtable) unable to find an inherited method for function ‘annotateWithFeatures’ for signature ‘"SimpleGRangesList", "list"’chromHMM_annotation not being converted from a list to a GRangesList.peakfile –> peakfiles to be consistent with other variables.badger with rworkfows:
precision_recall_matrixreport_timeoverlap_upset_plot:
UpSetR for ComplexUpsetto show percentages.heatmaply by checking args where it might be used:
check_heatmap_argstss_plot:
upstream=50
compute_corr:
bin_size = 200000 (takes <2s).predict_precision_recall
compute_corr and precision_recall now save outputs, including when run via EpiCompare Rmarkdown script.plot_precision_recall:
plot_precision_recall_prcurveplot_precision_recall_f1rebin_peaks:
sep between genomic coordinates in rownames.rebin_peaks:
drop_empty_chr to automatically drop chroms that aren’t in any of the peakfiles.intensity_cols in all relevant functions.translate_genome:
default_genome arg to handle genome=NULL.bpplapply:
BiocParallel across OS platforms.get_bpparam: Add args to allow users to choose which BiocParallel func to use.checkCache: Make default arg cache=BiocFileCache::BiocFileCache(ask = FALSE) to skip user input during runtime.precision_recall:
increment_threshold arg to n_threshold arg, using the seq(length.out=) feature to avoid accidentally choosing an inappropriately large increment_threshold.gather_files:
bpplapply.bpplapply.NULL.read_picard,read_multiqc,read_bowtie, read_trimgalore,read_bam,read_peaks
rbind_list.rebin_peaks/compute_corr: -Change defaultbin_size from 100 –> 5kb to improve efficiency and align with other defaults of other packages (e.g Signac).tss_plot:
EpiCompare: Pass up new args:
bin_sizen_thresholdworkersrebin_peaks unit tests.EpiCompare wasn’t being run when reference was a single unlisted GRanges object because it was indeed length>1, but the names were all NULL. Now fixed.plot_precision_recall: Set default initial_threshold= to 0.BiocParallel to parallel, as the former is extremely buggy and inconsistent.check_genome_build: Add translate_genome as prestep.rebin_peaks:
BiocParallel::bpmapply iterator.BiocParallel::bpmapply are of the same length, within the exact same bins, so that we can return just the bare minimum data needed to create the matrix (1 numeric vector/sample).rbinding the results and then casting them back into a matrix (which is safer bc it can handle vectors of different lengths), simply cbind all vectors into one matrix directly and name the rows using the predefined genome-wide tiles.rbinding a series of very long tables, this avoids the issue encountered here #103. This means this function is now much more scalable to many hundreds/thousands of samples (cells) even at very small bin sizes (e.g. 100bp).keep_chr allows users to specify whether they want to restrict which chromosomes are used during binning. By default, all chromosomes in the reference genome are used (keep_chr=NULL), but specifying a subset of chromosomes (e.g. paste0("chr",seq_len(12))) can drastically speed up compute time and reduce memory usage. It can also be useful for removing non-standard chromosomes (e.g. “chr21_gl383579_alt”, “chrUns…”, “chrRand…”).rebin_peaks now reports the final binned matrix dimensions and a sparsity metric.compute_corr:
reference to be NULL.EpiCompare with gather_files.compute_percentiles:
initial_threshold=0, so as not to assume any particular threshold.rebin_peaks:
plot_precision_recall: Don’t plot the reference as part of the PR curve.liftover_grl and added genome standardization.
get_chain_file.merge_all option.output_build options.dplyr.plyranges to Suggests.plot_precision_recall:
EpiCompare(precision_recall_plot=) param and documented.compute_consensus_peaks() as function for preprocessing peak files.
group_files() function to help assign each peakfile to a group based on substring searches.EpiCompare:
BiocParallel functions compatible with Windows.echo=FALSE instead of include=FALSE so the output messages will still be printed (without showing the code).here from Suggests.EpiCompare::EpiCompare.EpiCompare: accepts multiple reference files - creates individual reports for each reference. Added timing feature.save_output(): this function saves all plots and tables generated by EpiCompare. Also saves interactive heatmaps. Used in EpiCompare.Rmd.fig_length(): This function outputs dynamic figure height/width depending on the number of items. Used in EpiCompare.Rmd.prepare_reference: Validate reference input before passing to next step.genome_build to allow for different builds between reference and peaklist.blacklist to match GRanges list it’s being used to filter in tidy_peakfile.peaklist and reference.gather_files:
nf-core/cutandrun.return_paths to return only the paths without actually reading in the files.genome_build_output allows users to specify which genome build to standardise all inputs to.genome_build can now take a named list to specify different genome builds for peakfiles, reference, and blacklist.data.table to read/write tables.prepare_peaklist:
remove_empty to automatically drop any empty elements.check_list_names within.plot_chromHMM:
return_data.peaklist.output_dir creation recursive and without warnings.peaklist length check to prepare_peaklist.check_genomebuild: ensure necessary packages installed and that “genomebuild” is valid.check_cell_linesliftover_grlist: Dedicated liftover function, exported.checkCache.get_chromHMM_annotation can now take a list of cell lines as an argument.import_narrowPeak: Import narrowPeak files, with automated header annotation using metadata from ENCODE.gather_files: Automatically peak/picard/bed files and read them in as a list of GRanges objects.write_example_peaks: Write example peak data to disk.