Changelog • MotrpacRatTraining6mo

MotrpacRatTraining6mo 1.6.6 (2024-09-29)

Fix bug due to backwards incompatibility between ggraph 2.1.0 and 2.2.0. Require ggraph >= 2.2.0.

Point to development version of RCy3 from GitHub to avoid R-CMD check error RCy3: Can't install dependency uchardet.
Install plotrix from GitHub instead of CRAN to avoid R-CMD check warning Requires (indirectly) orphaned package: 'plotrix'. CRAN has marked plotrix as orphaned. plotrix is a dependency for mutoss, which is a dependency for metap.
Increment required MotrpacRatTraining6moData version.

Bug fix in transcript_timewise_da() to run with limited contrasts when tissue="VENACV".

Bug fix in atac_timewise_da() and atac_training_da() when n_features = Inf.

Accommodate RefMet IDs as feature IDs in plot_feature_normalized_data() and plot_feature_logfc().
Pass the user-supplied version of the feature identifier when plot_feature_logfc() is called within plot_feature_normalized_data().

For enrichment_network_vis() with argument return_html=TRUE, don’t require GNU sed.

In enrichment_network_vis(), include pathway names in determination of node labels.
In enrichment_network_vis(), use MotrpacRatTraining6moData::FEATURE_TO_GENE_FILT as feature_to_gene default instead of the unfiltered version.
Add more details to enrichment_network_vis() docs.

Add load_feature_annotation().
Add counts argument to transcript_normalize_counts() and atac_normalize_counts() to allow user-supplied data.
Add tissues, assays, and cluster arguments to enrichment_network_vis() to provide an alternate way for users to specify results from MotrpacRatTraining6moData::GRAPH_PW_ENRICH.
Add tests for enrichment_network_vis().
Use signif() instead of round() to display values in plot titles.
Add note about reproducibility issues to documentation for transcript_normalize_counts().
Move MotrpacRatTraining6moData from Imports to Depends. This means MotrpacRatTraining6moData is also attached (not just loaded) when MotrpacRatTraining6mo is attached.
Replace fetch_object(name_as_string) with .get, which internally uses get(name_as_string, envir=as.environment("package:MotrpacRatTraining6moData")).
Change URLs for load_methyl_feature_annotation() and load_atac_feature_annotation().
Fix bug in plot_feature_logfc() that prevented epigenetic features from being plotted.
Speed up plot_feature_logfc() for differential epigenetic features.
In get_rdata_from_url(), calculate log fold-change standard errors for METHYL differential analysis results: logFC_se = logFC/zscore

In plot_feature_normalized_data() and plot_feature_logfc(), handle input features that don’t exist in the data.

Dodge points in plot_feature_normalized_data() and plot_feature_logfc() for readability.
Make plot_feature_normalized_data() and plot_feature_logfc() more accommodating of features with multiple measurements.
Add add_adj_p parameter to plot_feature_normalized_data().
Add return_data parameter to plot_feature_logfc().
Return NULL instead of error if data doesn’t exist for plotting plot_feature_normalized_data() in loops.

Fix custom_cluster_pathway_enrichment() to accept a list of lists as input.
Use filtered feature-to-gene map to perform pathway enrichment.
In cluster_pathway_enrichment(), remove pathway enrichments driven by a single gene for consistency with the landscape paper.

Add df_to_numeric() to easily format data frames.
Add plot_feature_normalized_data() to plot normalized sample-level data for a single feature.
Add plot_feature_logfc() to plot differential analysis results for a single feature.
Add more examples and screenshots to the “Get Started” vignette.

Add functions to perform meta-analysis and meta-regression of redundant metabolites.
Add load_training_da() to load training summary statistics from GSC.
Retain order of input vial labels in result of viallabel_to_pid().

Use MotrpacRatTraining6moData::TRAINING_REGULATED_NORM_DATA and add a training_regulated_only argument to speed up plot_feature_trajectories()
Fix bugs in get_tree_plot_for_tissue() and get_trajectory_sizes_from_edge_sets()
Add scale argument to call_pca_outliers() (default TRUE)
Set default values for arguments plot and verbose in call_pca_outliers() (default TRUE)

Use a hashmap to speed up calculation of similarity scores in enrichment_network_viz() (thanks, Jay Yu)
Improve visInteraction settings in enrichment_network_viz(), i.e., show edges on zoom and slow down zoom
Don’t run memory-intensive examples and vignette code blocks to prevent GitHub Actions OOM error

First release