rogerginber / rhermes Goto Github PK

Hi there,

I'm testing this RHermes package, just following the RHermes_UserGuide.Rmd and use the example datafile attacthed to the package (MS1TestData.mzML). When I get to the step of find.SOI, it pops out an error:

s <- getSOIpar("double")
myHermes <- findSOI(myHermes, s, 1)
-------- Processing SOI 1 out of 1 --------

Starting density filtering:
Running Density Filter
Running Density Interpreter
Running Density Filter
Running Density Interpreter
Now Grouping:
Merging filter 2 out of 2 with previous SOI list
Adding entries unique from the first DF
Adding entries unique from the second DF
Solving inner conflicts, Round: 1
Initial peak retrieval:

Starting group characterization:
Mass calculation:
Shortening and selecting long groups:
Error in nrow(pks) : object 'pks' not found

how should I solve this problem?
Thanks!

Hi @RogerGinBer ,
By the time you explore your SOIs using RHermes or any other software, I think having the appex information (in seconds) for each entry in the SOIList would be a nice add on. What do you think?

Dear @RogerGinBer,

Regarding RHermes shiny app, the convention to name SOIs seems to be (SOI id number + formula adduct + intensity). I would appreciate adding retention time to the SOI names in SOI selector options within SOI exploration tab. This would help to quickly navigate SOIs

RHermes shiny app, SOI exploration window, Isotopic fidelity tab, I would suggest changing "Isotopic cosine score" by "Isotopic similarity Index" since the metric is no based in cosine anymore and might change in further versions

Hi Roger,

I've manually installed miniconda and tensorflow, which I believed the installation has been successful:

reticulate::use_condaenv("newenv", required = TRUE)
tensorflow::tf_config()
Loaded Tensorflow version 2.1.0
TensorFlow v2.1.0 (D:\Miniconda3\envs\newenv\lib\site-packages\tensorflow_init_.p)
Python v3.6 (D:/Miniconda3/envs/newenv/python.exe)
model <- keras::load_model_hdf5(system.file("extdata", "ImprovedModel.h5",

•                                         package = "RHermes"))
  

2022-01-04 20:23:23.206567: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2

is(model, "python.builtin.object")
[1] TRUE

I've downloaded the original data from ZENODO and attempted to go through all the process using RHermesGUI(). When I reached the step of SOI filtering - cleanup, the GUI turned grey and froze, and I got msg like this:

Listening on http://127.0.0.1:5091
Warning in readRDS(f) :
strings not representable in native encoding will be translated to UTF-8
Computing isotopic elution similarity:
Calculating isotope similarity:
Calculating isotopic fidelity metrics:
Removing confirmed isotope entries:
Recalculating peaklist for plotting:
Warning: Error in rep: invalid 'times' argument
2: shiny::runApp
1: RHermesGUI

Could you please let me know what's gone wrong and how to solve this issue?

Cheers,
Emily