The isoorbi from isoverse

isox needs to export more significant digits for the time.min column

currently it exports only 3 decimals (0.001) but some scans seem to record data on higher frequency. this leads to time stamps between consecutive scans that appear identical but are likely not

release isoorbi 1.2.0

implement analyze_shotnoise (data, compound, num, denom, ratio_method)
implement some shot noise plotting function
check documentation of all functions
update vignettes to include orbi_filter_flagged_data() and other new functions
rename orbi_calculate_ratio() to orbi_calculate_summarized_ratio()
#37
#38
implement a calculate_ratios function that doesn't do any grouping, just calculates ratios (and can be used standalone)without grouping - that way easy single ratio calculations are possible (make calculate_ratio() internal only) -> the summarize_results calls this but with grouping functions prepended
provide a function to easily backlay plot with the block definitions
show-case the workflow with Nils' flow injection scripts
maybe make a flow chart to illustrate how this works (consider including this in the paper)
support multi file read with orbi_read_isox()
convert the filter functions to orbi_flag_satellite_peaks(), orbi_flag_weak_isotopocules() and orbi_flag_outliers(intensity_window = c(2, 98)) as well as a orbi_remove_flagged_data() function

consider `define_blocks_from_table()`

define_basepeak() fails if satellite peaks have not been removed

> orbi_read_isox("paper/data/tfa/TFA_M0_1uscan_120kRes.isox") |> orbi_define_basepeak("M0")
orbi_read_isox() is loading .isox data from 1 files:
 - paper/data/tfa/TFA_M0_1uscan_120kRes.isox
orbi_define_basepeak() is setting the M0 isotopocule as the ratio denominator...                                                                                                                                                                                          
Error in `dplyr::mutate()` at isoorbi/R/utility_functions.R:440:2:
ℹ In argument: `basepeak_ions = .data$ions.incremental[.data$isotopocule == "M0"]`.
ℹ In group 3: `filename = TFA_M0_1uscan_120kRes`, `compound = TFA`, `scan.no = 3`.
Caused by error:
! `basepeak_ions` must be size 4 or 1, not 2.

This is because there are not the same number of basepeak ions for all peaks.

running it with satellite peak removal first works:

> orbi_read_isox("paper/data/tfa/TFA_M0_1uscan_120kRes.isox") |> orbi_filter_satellite_peaks() |> orbi_define_basepeak("M0")
orbi_read_isox() is loading .isox data from 1 files:
 - paper/data/tfa/TFA_M0_1uscan_120kRes.isox
orbi_filter_satellite_peaks() is removing minor signals (e.g., satellite peaks) that were reported by IsoX...                                                                                                                                                             
orbi_define_basepeak() is setting the M0 isotopocule as the ratio denominator...

allow ratio calculation for single scan

currently this is tricky because segmenting requires blocks definition first and ratio calculations require more than 1 values please chime in @brettdavidheiser how you would ideally like to have this implemented

Add a segmenting function

orbi_segment(n = INTEGER; by = c("scan", "time")) => add column "segment"

adjust other functions to use "segments" as a grouping variable

`orbi_summarize_results()` fails with new dplyr version

should factor_in_order be exported?

should this function be exported by isoorbi or should forcats instead be a suggested package in the DESCRIPTION (so we can use forcats::in_order() in vignettes? I'm leaning towards the latter but implemented the prior for now. What do you all think?

Speed up calculating data for shot noise plot

Release isoorbi 1.0

First release:

usethis::use_cran_comments()
Update (aspirational) install instructions in README
Proofread Title: and Description:
Check that all exported functions have @return and @examples
Check that Authors@R: includes a copyright holder (role 'cph')
Check licensing of included files
Review https://github.com/DavisVaughan/extrachecks

Prepare for release:

git pull
Check if any deprecation processes should be advanced, as described in Gradual deprecation
urlchecker::url_check()
devtools::check(remote = TRUE, manual = TRUE)
devtools::check_win_devel()
rhub::check_for_cran()
git push
Draft blog post

Submit to CRAN:

usethis::use_version('major')
devtools::submit_cran()
Approve email

Wait for CRAN...

orbi_summarize_results

It would be nice if you could pass more columns through the summarize function. For example I parsed the file name into additional columns and would like the parts to be within the summarized results. Another example is the mean of of the tic or mean of the ions.incremental if you want to make a plot showing the concentration vs ratio.

The work around right now is to summaries the results separately and then do a join call.

rename orbi_calculate_ratios?

Should this be renamed to orbi_calculate_ratio() as it does calculate a single value ratio?

create vignette for flow injections

orbi_plot_shot_noise() isotopocule filter

If you have lot of isotopocules the shot noise plots get very busy, especially when facet_wrap. It might be nice to have a built in filter like as in the fuction call orbi_plot_raw_data(isotopocules = c("13C"), y= ratio). Of course it can be done with a filter call before hand.
flageddata_w_ratios |> orbi_analyze_shot_noise() |> dplyr::filter(isotopocule %in% c("13C", "18O")) |> orbi_plot_shot_noise()

Flow chart in README is not showing on website (for me)

## Package layout

![](figure_flowchart.png)

using pkgdown::build_site()

Package as desktop app

orbi_adjust_block -time

If you set set_end_time.min to 0 or negative you get a dplyr::filter() error. Need to add check.

additional options for `orbi_define_block_for_flow_injection()`

we discussed that the following parameters would be useful for defining the data block in flow injenctions (in addition to the time window definitions) by basically providing a way to more automatically detect peak maxima:

intensity_setpoint = 95
intensity_cutoff = 10

--> this basically selects the peak by taking only everything above 10% of the 95% percentile of the data inside the time window
please correct me if that's wrong @NilsKuhlbusch

note that this is planned for v1.4 at the earliest

create vignette for shotnoise

implement plotting function(s) for flagged data

provide info messages of what happened in utility functions

so user can get a sense what happened - e.g. x% records filtered, etc.
could also be included in the simplify function

Loading multiple isox files with orbi_read_isox()

Release isoorbi 1.3.0 (CRAN requirements)

Prepare for release:

Submit to CRAN:

usethis::use_version('minor') - careful: this changes the version automatically one minor level up!
devtools::submit_cran()
Approve email

Wait for CRAN...

support start and end scan number in orbi_add_blocks_to_plot

currently only works for time.min x axes

catch errors not just warnings in function calls

use try_catch_all

Need smaller test file for dual inlet

Replace extdata/testfile_DualInlet.isox with a smaller version (e.g. 10 microscans) to meet CRAN file size limits (package total max 5MB)

Enable x axis in shot noise plot to show analysis time (in addition to n_effective)

rework utility functions to do grouping like summarize_results() does

several of the utility functions do default groupings that might be good to include as a paramter

shotnoise plot axis not passing to ggplotly correctly

When you plot the shot noise and then pass it to ggplotly for a dynamic graphs the axis become bogus. see attached photos. might be out of scope

implement `define_block(start, end, name, data_type = ?)`

speed optimization

some functions could probably be optimized a bit for speed - for example, orbi_define_basepeak() could potentially be implemented with a grouped mutate instead of a full join for speed increases

implement delta calculation function

for example something modelled after https://github.com/isoverse/isoprocessor/blob/812646ebab0faff38c30ace78be82a16704aed66/R/calculations.R#L221 except it can be for all ratios at once given the base peak structure

aim for v. 1.4? ideally for flow injections AND dual inlet samples

important decisions to make:

how to deal with segmented blocks' averaging and error calculation when bracketing sample/standard?

note issue #46 is also relevant for this as more automated sequence information is very useful to include

Plotting multiple ratios in a single shotnoise plot

Can we allow calculating and plotting multiple ratios in shot noise analysis?

Prepare release of isoorbi version 1.4 (CRAN requirements)

Once all new features are implemented and issues are addressed, prepare for release:

Submit to CRAN:

usethis::use_version('minor') - careful: this changes the version automatically one minor level up!
devtools::submit_cran()
Approve email

Wait for CRAN...

Release isoorbi 1.1.0

Prepare for release:

Submit to CRAN:

usethis::use_version('minor')
devtools::submit_cran()
Approve email

Wait for CRAN...

isoverse / isoorbi Goto Github PK

isoorbi's Introduction

isoverse

isoorbi's People

Contributors

Stargazers

Watchers

Forkers

isoorbi's Issues

Recommend Projects

Recommend Topics

Recommend Org