Skip to contents

Changelog

Source: NEWS.md

qSIP2 0.18

qSIP2 0.17

  • New functions to work with JGI spike-ins
  • Changed plot_feature_occurrence() to show all features by default
  • Print methods for the 4 main objects to show summary information
  • Added better error checking for sample data with duplicated sample names (#17)
  • Deprecated the plot_filter_gradient_position() function and replaced with plot_filter_results(), and updated vignettes

qSIP2 0.16

qSIP2 0.15

  • Introducing functions to work with multiple qSIP2 objects at once
  • Added run_comparison_groups() to launch multiple qSIP2 EAF workflows
  • Updated summarize_EAF_values() and plot_EAF_values() functions to work with multiple qSIP2 objects. Plotting puts each group in it’s own facet, and each facet is sorted individually by EAF values. The top argument also works for each group independently, but the confidence value is shared across all groups.
  • Less strict isotope validation to allow comparing mismatched labeled with unlabeled sources. For example, you can compare 18O labeled against 12C unlabeled sources.

qSIP2 0.14

  • Added grouping variable to get_N_total_it() to summarize by metadata variables
  • Fixed bug where N_total_it should be just from unlabeled, not labeled + unlabeled
  • Added linear growth model in addition to the existing exponential model. Exponential is still default.
  • Renamed unlabeled and labeled to N_light_it and N_heavy_it, respectively
  • Growth calculations more finalized
  • when recalculating N_light_it, qSIP2 now recalculates N_heavy_it to keep N_total_it constant

qSIP2 0.13

  • Beta functions for growth calculations
  • Added time and total_abundance to qsip_source_data() as required arguments if you want to do the growth calculations
  • Added calculate_time_zero_abundance() to summarize the time zero abundance for each feature
  • Added run_growth_calculations() to calculate growth (r), birth (b) and death (d) rates from total abundances and EAF values
  • Added summarize_growth_values() to summarize r, b and d rates
  • Added plot_growth_rates() to visualize growth r, b and d rates
  • Two different growth plot types based on either rates or N copies
  • Both timepoints (time zero and time t) are now reported during growth calculations
  • Other features that make comparing rates between timepoints that are not time zero easier
  • Added a group argument to run_feature_filter() to embed a grouping name to the qsip object
  • Ability to adjust the total abundance copies using a per-row volume adjustment in the source data
  • Moved some Koch, 2018 equations to their own functions
  • Resampling now calculates r_net and N_total_it values

qSIP2 0.12

qSIP2 0.11

  • Fixed run_resampling() to not get confused when using sample names that are integers.
  • Fixed example_source_df and example_sample_df to remove the built in validation errors (missing isotoplog in the source data, and fractions as characters in the sample data). These dataframes are now valid objects for the package
  • Added validation checks for existing standard names in dataframes. For example, if trying to use a data.frame with source data that already has a source_mat_id column, but you designate a different column as the source_mat_id, it will throw an error. This is to prevent column name collisions and potential unintended consequences.
  • Added internal function validate_standard_names() to check for existing standard names in dataframes.
  • Added alpha function plot_resampling_convergence() to follow when the CoV of the resamplings converge to a stable value
  • Added new vignette("resampling") for more details about the resampling procedure
  • Fixed run_EAF_calculations() to work with allow_failures logic
  • Add plot_feature_resamplings() to plot resampling results for a single or list of feature_ids
  • Introduced ability to keep only successful resampling attempts, and discard failures. This keeps run_resampling() from failing if the sample count is low, but could result in feature_ids with less than the expected number of resamples. This is controlled by the allow_failures argument in run_resampling(). Two functions have been added to inspect the results of resampling: get_resample_counts() and plot_successful_resamples().

qSIP2 0.10

  • Added infer_source_data() function to infer source data from a sample data data frame
  • Update documentation of internal functions
  • Fixed validate_gradient_pos_density() (and tests) to not fail with bulk data that has a gradient_position of -1 (#8).
  • Fixed validate_isotopes() to accept standard unfractionated terms like “bulk” or “time0” so they bypass isotope validation.
  • plot_sample_curves() and plot_source_wads() have been updated to ignore unfractionated samples/sources
  • Removed requirement for gradient_pos_rel_amt column in the imported sample dataframe. If you have one you can still pass the column name. If you don’t, it will run the add_gradient_pos_rel_amt() silently using the gradient_pos_amt column, and provide a message that it is doing so.
  • Updated vignette("feature_data")
  • Documentation for get_dataframe()
  • Started a NEWS.md file to keep track of changes
  • Updated plot_sample_curves() and plot_source_wads() to use existing WAD values that were pre-calculated when making the qsip_data object. This means they now require a qsip_data object as input and no longer accept a sample or source object. The main workflow vignette was updated to reflect these changes.
  • Renamed data() to get_dataframe() to match the naming scheme of other functions.