Page 46 - Read Online

Wang et al. Microbiome Res Rep 2024;3:39 https://dx.doi.org/10.20517/mrr.2024.21 Page 11 of 17

Figure 3. Quantitation of small proteins and antimicrobial peptides in mouse fecal samples. (A) Count and (B) relative abundance of
small proteins in samples using a 100 amino acid cut-off. Functional enrichment analysis of identified small proteins in (C) the DDA-
PASEF dataset and (D) the DIA-PASEF dataset, using all identified proteins as background in DDA and DIA datasets, respectively.
Functional annotation was performed using GhostKOALA, and only significantly enriched functions were shown (adjusted P value ≤
0.05). (E) Count and (F) relative abundance of AMPs when searched against the AMPsphere database. Facet_grid function in ggplot2
was used to generate different sections; DC and DDA groups are on the left side, while NC and DIA on the right side. DDA: Data-
dependent acquisition; PASEF: parallel accumulation-serial fragmentation; DIA: data-independent acquisition; AMPs: antimicrobial
peptides; DC: differential centrifugation.

workflows, and quantification using DIA-PASEF presented the highest number of identifications and
relative abundance of small proteins within the sample [Supplementary Figure 6].

Small proteins are implicated in diverse functions of the microbiome, including defense against other
microbes or pathogens . Among the predicted small proteins in the human microbiome, around 30% were
[26]
predicted to be secreted or transmembrane proteins and 39 protein families were predicted to be novel
[25]
AMPs . By using a deep learning approach, Ma et al. identified 2,349 potential AMPs from human
[27]
microbiome metagenomic data and preliminary biological validation showed > 80% positive rate . More
recently, Santos-Júnior et al. leveraged a vast dataset of more than 1.5 million metagenomes or microbial