bogus_py

Overview

Welcome to the Bristol Organic Geochemistry Unit System - Python (BOGUS Py for short).

This is a Python package developed by the Organic Geochemistry Unit to help process isotope ratio mass spectrometry (IRMS) data. This is done in two ways:

Data preparation

Our IRMS instruments (particularly the IsoPrime 100) produce horrible proprietary data files, requiring each run to be opened and the data manually extracted before we can calculate δ values. With the data preparation functions in this package one should be able to create one long .csv or Excel file from a folder full of data files without any copy-pasting at all!

We currently have data preparation functions for the following instrument files:

• Delta V (.dxf)
• Delta XP (.dxf)
• IsoPrime 100 (.raw folders; these are particularly annoying)

Note that processing of .dxf files from the Delta instruments requires an extra step, see below.

Data analysis

Once δ values have been obtained, we also need to perform corrections like scale normalisation. We also have functions for that!

Installation

This package can be installed into a Python environment with pip, using the following command:

pip install git+https://github.com/ogubristol/bogus_py.git

To submit modifications/additions, please branch and create a merge request. For help with this, contact Nick Hall.

DXF Pre-processing

Currently, this package cannot directly handle output files from Thermo IRMS software (e.g. .dxf files). Handily, the isoreader R package has been developed to handle these files, and can export useful data as easy-to-read .csv files, which this package can then process.

For a directory (folder) containing .dxf files, we can create single .csv file containing every run's data as follows:

library(isoreader)

DIR <- "path/to/directory/data"
OUTFILE <- "path/to/directory/all_run_data.csv"

OUTFILE <- paste(DIR, OUTFILE, sep = "")

dxf_files <- list.files(path=DIR, pattern="*.dxf", full.names=TRUE, recursive=FALSE)
file_data <- iso_read_continuous_flow(dxf_files)
output_data <- file_data |> iso_get_vendor_data_table(
    include_file_info=c(file_datetime, starts_with("Identifier"))
    )

write.table(output_data, OUTFILE, sep=",", row.names=FALSE)

Where the .csv contains the "vendor data" (the per-peak isotope ratio data that you'd ordinarily copy and paste into an Excel sheet) for each run as well as some extra information like the date/time the run was started.

This .csv is now ready to be processed in Python with this package.