Laudicella 2019 Lipidomics analysis of juveniles blue mussels

Published: 30 January 2020| Version 1 | DOI: 10.17632/w57zy87s68.1
Contributor:
vincenzo laudicella

Description

This dataset includes the raw positive and negative LC-MS data of blue mussel spat subjected to five different diets: single strains diet of Cylindrotheca fusiformis 1017/2 – CYL, Isochrysis galbana 927/1– ISO, Monodopsis subterranean 848/1 – MONO, Nannochloropsis oceanica 849/10– NANNO and a commercial algae paste –SP), and natural growth-out (OUT). Spat were fed with the above-reported diets for a total of 4 weeks. At the end of the trial, spat were instantly frozen, freeze-dried and pulverised. Lipid extracts were done according to Folch technique. Data were acquired in ESI POS and NE with a high-resolution LC-MS platform (Exactive Thermo). The data are reported as precursor ion (MS') mass spectra. The mass error was kept below 5 ppm by routine calibration of the mass spectrometer. Lipids were separated with a reverse-phase column (C18Hypersyl Gold 100x2.1 mm 1.9nm particle size, ThermoFisher) kept at 50 ⁰C. Details of separation protocol are provided with the companion publication. Data were automatically integrated with Excalibur software 4.1 (Thermo). The diets had a diverse fatty acid profile which influenced the lipidome of the spat. Good performing diets increased triglycerides (TG) content of spat. Diets also influenced the polar lipid composition of spat, with changes on the principal membrane phospholipids (PC and PE).

Files

Steps to reproduce

1) Process data via Progenesis QI software (Nonlinear Dynamics, Waters) 2) Chromatograms were automatically aligned using a QC as a reference point. 3) Peak picking and deconvolution were completed following automatic settings of the software, with intensity threshold of 1xE5 and 1xE4 respectively for POS and NEG ionization modes. 4) Data were normalised according to the total ion current of each chromatogram. 5)Main lipid adducts for lipid research reported in S1 Table --> Original publication 6) Lipid identification was achieved by searching the lipid dataset versus LIPID MAPS (www.lipidmaps.org), HMDB (www.hmdb.ca), Metlin (www.metlin.scripps.edu) 7) Resulting peak intensity table (PIT) processed via MetaboAnalystR --> www.metaboanalyst.ca 8)Upload the PIT and filter: - 30% of missing values and substitution of remaining missing values with a small value (half of the minimum intensity value). - Features with low repeatability or low constant values were filtered out using QC samples (20% variation) and inter-quantile range - data was then scaled via Pareto scaling 9)run multivariate and univariate statistics