We developed an R package called MetabImpute to explore various aspects of metabolomics data sets, especially those with technical or biological replicates. Included in the package are tools to evaluate the mechanisms of data missingness and variable distributions, tools to impute missing values using several methods, and to evaluate the effects of imputation on data reproducibility, measured by intra-class correlation.
https://github.com/BeanLabASU/MetabImpute
The paper describing MetabImpute is published here:
Trenton J. Davis*, Tarek Firzli*, Emily A. Higgins Keppler*, Matthew Richardson, Heather D. Bean. (2022) Addressing missing data in GC×GC metabolomics: Identifying missingness type and evaluating the impact of imputation methods on experimental replication. [*co-first authors] Analytical Chemistry, 94, 10912-10920. Published: Link Preprint: Link
We make all of our published metabolomics data available to the public through the Metabolomics Workbench. Brief descriptions and links to our data sets are provided, below.
Study Title: Pseudomonas aeruginosa cystic fibrosis clinical isolates volatile metabolomics
Description: Untargeted metabolomics analysis of in vitro headspace volatiles from 81 Pseudomonas aeruginosa bacterial isolates from individuals with cystic fibrosis. Headspace volatiles were collected using solid-phase microextraction (SPME) (in triplicate) and comprehensive two-dimensional gas chromatography and time-of-flight mass spectrometry (GCxGC-TOFMS). 15 replicates of un-inoculated media were prepared and analyzed in parallel, for a total of 258 samples.
Study ID: ST001414
https://doi.org/10.21228/M89Q4F
Study Title: Identify putative volatile biomarkers of Coccidioides spp. grown in vitro
Description: GCxGC-TOFMS untargeted metabolomics analysis of in vitro headspace volatiles from 6 Coccidioides posadasii and 6 Coccidioides immitis strains grown in triplicate under conditions to induce two growth phases (spherule and mycelia). A total of 72 samples and 6 blanks were analyzed.
Study ID: ST001659
http://dx.doi.org/10.21228/M85H6W
Study Title: Identify putative volatile biomarkers of Valley fever using a murine lung infection model
Description: GCxGC-TOFMS untargeted metabolomics analysis of headspace volatiles from the bronchoalveolar lavage fluid of 6 mice infected with Coccidioides posadasii, 6 mice infected with Coccidioides immitis, and 4 PBS controls. Samples were analyzed in technical replicates.
Study ID: ST002350
http://dx.doi.org/10.21228/M85H6W
Study Title: Influence of growth medium on the volatilomes of Pseudomonas spp. and Staphylococcus spp.
Description: GCxGC-TOFMS untargeted metabolomics analysis of in vitro headspace volatiles from Staphylococcus aureus, Staphylococcus epidermidis, Pseudomonas aeruginosa strains PAO1 and PA14, and Pseudomonas chlororaphis cultured in triplicate in LB, TSB, BHI, and MHB media. A total of 72 samples were analyzed.
Study ID: ST001426
https://dx.doi.org/10.21228/M88Q44