Van-Krevelen-diagrams-R

General description

Here is the plot for a nice visualisation of Van Krevelen diagrams to visualise mass spectrometry data of natural organic matter (NOM) analysis. Van Krevelen diagrams are used to characterise individual components of complex mixtures, NOM including. In its classic variant, it is a plot of atomic O/C versus atomic H/C from elemental analysis.

In this repo, you can find two plotting functions for both continuous and discrete values and the function for plotting the density of diagram saturation by molecules that can be used for the following assigning to precursor groups.

The data

To test functions and to understand how to apply it to the data, a test dataset is included.

	Column	Non-Null Count	Dtype
0	file_id	9146 non-null	object
1	mz	9146 non-null	float64
2	i_magnitude	9146 non-null	int64
3	m	9146 non-null	float64
4	m_cal	9146 non-null	float64
5	ppm	9146 non-null	float64
6	nm	9146 non-null	int64
7	mf	9146 non-null	object
8	dbe	9146 non-null	int64
9	c	9146 non-null	int64
10	h	9146 non-null	int64
11	n	9146 non-null	int64
12	o	9146 non-null	int64
13	p	9146 non-null	int64
14	s	9146 non-null	int64
15	oc	9146 non-null	float64
16	hc	9146 non-null	float64
17	nc	9146 non-null	float64
18	sc	9146 non-null	float64
19	kmd	9146 non-null	float64
20	s_n	9146 non-null	float64
21	dbe_o	9146 non-null	int64
22	n_occurrence	9146 non-null	int64
23	n_assignments_orig	9146 non-null	int64
24	bp	9146 non-null	int64
25	rel_int	9146 non-null	float64
26	sum_int	9146 non-null	int64
27	rel_sum_int	9146 non-null	float64
28	ai	9146 non-null	float64
29	ai_category	9146 non-null	object
30	ai_mod	9146 non-null	float64
31	ai_mod_category	9146 non-null	object
32	form	9146 non-null	object
33	name	9146 non-null	object

Main columns

• file_id, name – refers to the names of samples in test-data
• c, h, n, o, p, s - amount of atoms in molecule of individual component; oc, hc, nc, sc – ratios of number of atoms in molecule;
• m – molecular mass;
• mf - molecular formula of molecule;
• n_occurrence – frequency of molecule occurrence;
• form – simple formula of molecule;
• ai, ai_mod – aromaticity indices calculated according to the following equations (Zherebker et al., 2022):

${AI} = \frac{1+C-O-S-0.5N-0.5H}{C-O-N-S} $

${AI_{mod}} = \frac{1+C-(СООН)_y-0.5N-0.5H}{C-(СООН)_y-N-S} $

• ai_category and ai_mod_category – categorisation of aromaticity indices according to the rules

• >= 0.5 and < 0.67 – aromatic structures
• > 0.67 – condensed structures
• < 0.5 – other structures

Functions

1. plotting functions for continuous and discrete values contains arguments:

• data – your mass spectrometry data,
• x – O/C ratio,
• y – H/C ratio,
• color – any target characteristic you choose for the third dimension to visualise.

2. plotting function for density mapping contains only data, x, y parameters, described above.

You can use functions to one sample dataset and to multisample long data, specifying sample names in the column name and adding facet_grid option to the function.

Visual examples