The metaproteomics field has recently gained more and more interest as a valuable tool for studying both the taxonomy and function of microbiomes, including those used in food fermentations. One crucial step in the metaproteomics pipeline is selecting a database to obtain high-quality taxonomical and functional information from microbial communities. One of the best strategies described for building protein databases is using sample-specific or study-specific protein databases obtained from metagenomic sequencing. While this is true for high-diversity microbiomes (such as gut and soil), there is still a lack of validation for different database construction strategies in low-diversity microbiomes, such as those found in fermented dairy products where starter cultures containing few species are used. In this study, we assessed the performance of various database construction strategies applied to metaproteomics on two low-diversity microbiomes obtained from cheese production using commercial starter cultures and analyzed by LC-MS/MS. Substantial differences were detected between the strategies, and the best performance in terms of the number of peptides and proteins identified from the spectra was achieved by metagenomic-derived databases. However, extensive databases constructed from a high number of available online genomes obtained a similar taxonomical and functional annotation of the metaproteome compared to the metagenomic-derived databases. Our results indicate that, in the case of low-diversity dairy microbiomes, the use of publically available genomes to construct protein databases can be considered as an alternative to metagenome-derived databases.

中文翻译：

---

低多样性乳制品微生物组鸟枪法宏蛋白质组数据库选择

宏蛋白质组学领域最近作为研究微生物组（包括食品发酵中使用的微生物组）的分类和功能的宝贵工具而受到越来越多的关注。宏蛋白质组学流程中的一个关键步骤是选择一个数据库，以从微生物群落中获取高质量的分类学和功能信息。构建蛋白质数据库的最佳策略之一是使用从宏基因组测序获得的样本特异性或研究特异性蛋白质数据库。虽然对于高多样性微生物组（例如肠道和土壤）来说确实如此，但对于低多样性微生物组中的不同数据库构建策略仍然缺乏验证，例如在发酵乳制品中发现的那些，其中含有很少物种的发酵剂培养物用过的。在这项研究中，我们评估了应用于宏蛋白质组学的各种数据库构建策略的性能，这些策略对使用商业发酵剂生产奶酪时获得的两种低多样性微生物组进行了分析，并通过 LC-MS/MS 进行了分析。检测到这些策略之间存在显着差异，并且通过宏基因组衍生数据库实现了从光谱中识别的肽和蛋白质数量方面的最佳性能。然而，与宏基因组衍生的数据库相比，由大量可用在线基因组构建的广泛数据库获得了类似的宏蛋白质组分类学和功能注释。我们的结果表明，在乳制品微生物组多样性低的情况下，使用公开的基因组构建蛋白质数据库可以被视为宏基因组衍生数据库的替代方案。