Description
Introduction
Beer is one of the oldest and most widely consumed beverages in the world. It contains a complex mixture of proteins from several organisms including plants (barley, hops, rice, wheat) and yeast, depending on the beer. The Proteomics Research Group (PRG) in the Association of Biomolecular Resource Facilities brought together an international consortium of proteomics laboratories (69 laboratories from 33 countries) to perform beer proteomics. An aliquot of specially brewed beer from the UC Davis Brewing Program was sent to 52 labs, and participants were encouraged to also use a widely available commercial beer (Heineken) and any other beer. In addition to helping connect scientists in trying times, we have also demonstrated the utility of multi-center studies with accessible materials.
Methods
Control beer brewed at UC Davis was shipped to participating laboratories at room temperature. Where shipping proved impractical, a widely available commercial beer was suggested as an alternative (Heineken). The PRG suggested a general digestion method, approximately 1 h (± 15 min) LC gradient, and using data-dependent acquisition. The suggested method consisted of precipitation, reduction, alkylation, and digestion with trypsin or LysC plus trypsin, but participants could use other methods. Raw mass spectrometry data (357 injections) and methods were deposited to MassIVE (MassIVE MSV000088080). Database searching was performed with MetaMorpheus against five species databases (barley, hops, rice, wheat, yeast) and contaminants. Mass tolerances were set automatically, carbamidomethylation was fixed, oxidized methionine was variable, and a 1 % FDR identification cutoff.
Preliminary Data
Of the 69 laboratories that signed up for the study, 35 returned data. Of these, 32 analyzed the PRG Beer and 17 analyzed Heineken (14 analyzed both). When shipping of the PRG Beer was prohibitive, participants were encouraged to use Heineken due to its global availability and perceived quality control. Participants were also encouraged to analyze other beers of their choosing, and 79 other beers were analyzed. In total there were 357 beer injections on 13 different types of mass spectrometers around the world. On average, 753.1 proteins were identified using MetaMorpheus, with the most being 2907 identified in the PRG Beer with a Thermo QE classic. Though database searching used non-UniProtKB non-RefSeq databases for barley, hops, and wheat, a downstream orthology conversion with BLAST found that despite their small size, the UniProtKB databases of these species adequately describe the mass specomtery data. Next, the 50 most abundant proteins identified in the PRG Beer and Heineken will be used for unsupervised clustering to determine how well beer proteomics performs across the world using the same or very similar beer. Finally, we will use the compareMS2 software tool to cluster data between beers and labs in an ID-free method. Though these studies focused only on the main grain additives to yeast, future work will expand the search space to include microorganisms that are part of the brewing process in some beers (i.e., bacteria in sours).
Novel Aspect
Multi-center proteomic studies using accessible materials, such as beer, can be successful, generate valuable results, and facilitate community building.
Beer is one of the oldest and most widely consumed beverages in the world. It contains a complex mixture of proteins from several organisms including plants (barley, hops, rice, wheat) and yeast, depending on the beer. The Proteomics Research Group (PRG) in the Association of Biomolecular Resource Facilities brought together an international consortium of proteomics laboratories (69 laboratories from 33 countries) to perform beer proteomics. An aliquot of specially brewed beer from the UC Davis Brewing Program was sent to 52 labs, and participants were encouraged to also use a widely available commercial beer (Heineken) and any other beer. In addition to helping connect scientists in trying times, we have also demonstrated the utility of multi-center studies with accessible materials.
Methods
Control beer brewed at UC Davis was shipped to participating laboratories at room temperature. Where shipping proved impractical, a widely available commercial beer was suggested as an alternative (Heineken). The PRG suggested a general digestion method, approximately 1 h (± 15 min) LC gradient, and using data-dependent acquisition. The suggested method consisted of precipitation, reduction, alkylation, and digestion with trypsin or LysC plus trypsin, but participants could use other methods. Raw mass spectrometry data (357 injections) and methods were deposited to MassIVE (MassIVE MSV000088080). Database searching was performed with MetaMorpheus against five species databases (barley, hops, rice, wheat, yeast) and contaminants. Mass tolerances were set automatically, carbamidomethylation was fixed, oxidized methionine was variable, and a 1 % FDR identification cutoff.
Preliminary Data
Of the 69 laboratories that signed up for the study, 35 returned data. Of these, 32 analyzed the PRG Beer and 17 analyzed Heineken (14 analyzed both). When shipping of the PRG Beer was prohibitive, participants were encouraged to use Heineken due to its global availability and perceived quality control. Participants were also encouraged to analyze other beers of their choosing, and 79 other beers were analyzed. In total there were 357 beer injections on 13 different types of mass spectrometers around the world. On average, 753.1 proteins were identified using MetaMorpheus, with the most being 2907 identified in the PRG Beer with a Thermo QE classic. Though database searching used non-UniProtKB non-RefSeq databases for barley, hops, and wheat, a downstream orthology conversion with BLAST found that despite their small size, the UniProtKB databases of these species adequately describe the mass specomtery data. Next, the 50 most abundant proteins identified in the PRG Beer and Heineken will be used for unsupervised clustering to determine how well beer proteomics performs across the world using the same or very similar beer. Finally, we will use the compareMS2 software tool to cluster data between beers and labs in an ID-free method. Though these studies focused only on the main grain additives to yeast, future work will expand the search space to include microorganisms that are part of the brewing process in some beers (i.e., bacteria in sours).
Novel Aspect
Multi-center proteomic studies using accessible materials, such as beer, can be successful, generate valuable results, and facilitate community building.
| Date made available | 2021 |
|---|---|
| Publisher | ZENODO |