This dataset presents raw data and intensities of metabolites detected in broncho-alveolar lavages (BAL) collected from children presenting severe asthma (n=20) and age-matched disease control children (n=10). Untargeted metabolomics was performed using liquid chromatography coupled with high resolution mass spectrometry (LC-HRMS) following an established workflow for sample preparation, data acquisition and treatment (DOI: 10.1016/j.jchromb.2014.04.025). LC-HRMS was performed on an Ultimate 3000 chromatographic system coupled to a Q-Exactive mass spectrometer (Thermo Fisher Scientific, Courtaboeuf, France) fitted with an electrospray source operating in the positive ionization mode (ESI+). Ultra-high performance LC (UHPLC) separation was performed using C18 column (Hypersil GOLD C18, 1.9 µm, 2.1 mm × 150 mm column, Thermo Fisher Scientific). Raw data was converted to .mzXML format using MSconvert (“ProteoWizard”, version 3.0.21079), and data extraction was performed using XCMS R package (https://workflow4metabolomics.org/), then providing one raw data file per sample (CLASSE_METABOLOME_RAW_DATA.zip). Raw data from quality controls (“QC”), corresponding to mix of equal volume from all samples that were analysed every 6 randomized samples, from QC dilutions, from buffer only (“Blanc”; injected 5 times at the very beginning and once at the end of sequence analysis) and from an extraction blank (sterile NaCl 0.9 % extracted and treated as samples), are also provided : all those controls are used for data normalization/standardization purposes. After filtration based on three quality criteria (phenomis R package, version 1.0.2), metabolites were annotated using an internal spectral database and confirmed by MS/MS experiments (DOI: 10.1021/ac300829f), allowing the highest confident level of identification (DOI: 10.1007/s11306-007-0082-2). In total, we identified 88 metabolites in BAL fluids (Intensities_annotated_metabolites_BAL.xlsx), which intensities in individual samples are provided as a first table. Metabolites characteristics are provided in a separate table (mass to charge ratio (m/z), retention time (rt), KEGG identifier and class of annotated metabolites). Our aim is to identify a local signature of severe asthma by conducting comprehensive multi-omics analysis of BALs from children with severe asthmatic versus non-asthmatic controls. Corresponding multi-omics data will be described in a data paper under submission, also describing all corresponding metadata.