Step 1: Load a data file¶

Tip

It is possible to try BioPAN using a demonstration file. Select the small dataset to understand how BioPAN works and the complete one to perform a more complex analysis 1.

How to prepare your data¶

BioPAN takes an input file in CSV (Comma-Separated Values) format containing quantitative data. The file structure should be as follows:

The first row contains sample labels (e.g.: ‘wild_type_1’, ‘wild_type_2’, ‘control_1’, ‘control_2’). The file must contain at least two conditions (e.g.: ‘wild_type and ‘control’) and two samples per condition.

The first column contains lipid molecular species. The lipid subclasses recognised by BioPAN are provided in the table below.

Columns 2 to n contain molecular concentration quantification.

Abbreviation in BioPAN	Subclass in LIPID MAPS
MG	Monoacylglycerols [GL0101]
DG	Diacylglycerols [GL0201]
TG	Triacylglycerols [GL0301]
PA	Diacylglycerophosphates [GP1001]
PC	Diacylglycerophosphocholines [GP0101]
PE	Diacylglycerophosphoethanolamines [GP0201]
PS	Diacylglycerophosphoserines [GP0301]
PG	Diacylglycerophosphoglycerols [GP0401]
PI	Diacylglycerophosphoinositols [GP0601]
PIP	Diacylglycerophosphoinositol monophosphates [GP0701]
PIP2	Diacylglycerophosphoinositol bisphosphates [GP0801]
PIP3	Diacylglycerophosphoinositol trisphosphates [GP0901]
LPA	Monoacylglycerophosphates [GP1005]
LPC	Monoacylglycerophosphocholines [GP0105]
LPE	Monoacylglycerophosphoethanolamines [GP0205]
LPS	Monoacylglycerophosphoserines [GP0305]
LPG	Monoacylglycerophosphoglycerols [GP0405]
LPI	Monoacylglycerophosphoinositols [GP0605]
O-DG	1-alkyl,2-acylglycerols [GL0202]
O-PA	1-alkyl,2-acylglycerophosphates [GP1002]
O-PC	1-alkyl,2-acylglycerophosphocholines [GP0102]
O-PE	1-alkyl,2-acylglycerophosphoethanolamines [GP0202]
O-LPA	Monoalkylglycerophosphates [GP1006]
O-LPC	Monoalkylglycerophosphocholines [GP0106]
O-LPE	Monoalkylglycerophosphoethanolamines [GP0206]
P-PC	1-(1Z-alkenyl),2-acylglycerophosphocholines [GP0103]
P-PE	1-(1Z-alkenyl),2-acylglycerophosphoethanolamines [GP0203]
P-LPA	1Z-alkenylglycerophosphates [GP1007]
P-LPC	1Z-alkenylglycerophosphocholines [GP0107]
P-LPE	1Z-alkenylglycerophosphoethanolamines [GP0207]
CL	Diacylglycerophosphoglycerophosphodiradylglycerols [GP1201]
Cer	N-acylsphingosines (ceramides) [SP0201]
dhCer	N-acylsphinganines (dihydroceramides) [SP0202]
SM	Ceramide phosphocholines (sphingomyelins) [SP0301]
dhSM	Ceramide phosphocholines (sphingomyelins) [SP0301]
SPB	Sphing-4-enines (Sphingosines) [SP0101]
dhSPB	Sphinganines [SP0102]
SPBP	Sphingoid base 1-phosphates [SP0105]
dhSPBP	Sphingoid base 1-phosphates [SP0105]
FA	Fatty acids [FA0101]
FaCoA	Fatty acyl CoAs [FA0705]

Note

The subclass term is used to refer to the lipid names in the above table and the molecular species one is used to refer to the complex subclass/structure. The structure corresponds to the number of carbons and bonds separated by a colon. For example the ‘DG’ subclass and the ‘DG 30:0’ molecular species.

Following is an example of an input file:

sample,condition1_1,condition1_2,condition2_1,condition2_2
FaCoA(16:1),87.86,146.49,1041.98,1082.65
FaCoA(17:1),71.98,132.04,354.17,399.86
FA(16:0),25006.43,25477.98,32934.67,33290.69
FA(18:0),19293.23,21057.76,19763.06,20423.42
MG(16:0),28.27,29.62,68.35,66.21
MG(16:1),2.19,2.77,5.00,4.78
MG(18:0),61.31,66.28,92.66,94.77
MG(18:1),67.79,60.46,86.52,91.67
MG(18:2),26.96,24.59,36.34,30.88
MG(20:4),0.53,0.84,2.59,4.95
MG(22:6),0.17,0.49,0.53,0.15
DG(30:0),206.01,159.52,497.23,550.29
DG(30:1),109.54,93.64,246.66,228.54
DG(32:0),1539.57,1764.44,3237.65,3941.12
DG(32:1),1954.26,1976.83,2972.57,2845.93
DG(32:2),575.92,632.22,1216.70,1114.80
DG(34:0),98.05,118.36,649.18,910.87
TG(46:3),1.69,4.28,1.77,0.85
TG(46:4),0.19,0.52,0.27,0.16
TG(48:0),4.72,16.70,11.85,4.09
TG(48:1),30.94,65.80,40.25,21.59
TG(48:2),36.42,79.39,40.63,19.11
TG(48:3),16.73,41.24,14.45,6.31
O-PC(32:0),13.99,18.52,35.82,21.02
O-PC(32:1),10.25,10.70,25.98,12.10

It is also possible to download the demonstration files 1 to understand the required structure.

Note

After loading a dataset, LipidLynxX 2 is launched in the background. LipidLynxX takes the input file and converts the nomenclature of the lipid molecular species to BioPAN nomenclature. It will also equalise the structure of the lipid molecular species to the Bulk level if it not already (e.g.: PE(34:1)).

Warning

Currently, BioPAN:

Does not analyse oxidised lipids.
Only consider the sphingoid bases with 18 carbons and 0 (dhCer/dhSM) or 1 double bond (Cer/SM) for the Sphingolipids. For example: it means that the lipids molecular species ‘Cer 24:0’ corresponds to ‘Cer(d18:1/24:0)’ and ‘dhCer 20:4’ corresponds to ‘Cer(d18:0/20:4)’ (same for SM/dhSM).

Data summary¶

The data summary table summarises the data extracted from the imported file:

Unrecognised species: molecular species whose subclass is not recognised by BioPAN. The recognised lipid subclasses are available in the table from the section How to prepare your data.

Processed species: molecular species that are involved in at least one reaction (read more: Which lipids and reactions appeared on the graph?).

Unprocessed species: molecular species that are not involved in any reactions.

It is possible to see the list of molecular species “unrecognised”, “processed” and “unprocessed” by clicking on the associated numbers. For the categories “processed” and “unprocessed”, you can also visualise the nomenclature used by BioPAN from your input dataset.

Note

A distinction is made between fatty acids and lipids because BioPAN stores reactions in a different way. It stores the reactions at the level of the molecular species for FA (example FA (16: 0) -> FA (16: 1)) and at the level of the subclass for lipids (example DG -> PA).

1(1,2): Oliver Hahn, Lisa F. Drews, An Nguyen et al. A nutritional memory effect counteracts the benefits of dietary restriction in old mice. Nat Metab 1, 1059–1073 (2019).
2: Zhixu Ni, Maria Fedorova. LipidLynxX: a data transfer hub to support integration of large scale lipidomics datasets. bioRxiv, 2020.04.09.033894.