Generation of functional oligopeptides that promote osteogenesis based on unsupervised deep learning of protein IDRs.

Deep learning (DL) is currently revolutionizing peptide drug development due to both computational advances and the substantial recent expansion of digitized biological data. However, progress in oligopeptide drug development has been limited, likely due to the lack of suitable datasets and difficulty in identifying informative features to use as inputs for DL models. Here, we utilized an unsupervised deep learning model to learn a semantic pattern based on the intrinsically disordered regions of ~171 known osteogenic proteins. Subsequently, oligopeptides were generated from this semantic pattern based on Monte Carlo simulation, followed by in vivo functional characterization. A five amino acid oligopeptide (AIB5P) had strong bone-formation-promoting effects, as determined in multiple mouse models (e.g., osteoporosis, fracture, and osseointegration of implants). Mechanistically, we showed that AIB5P promotes osteogenesis by binding to the integrin α5 subunit and thereby activating FAK signaling. In summary, we successfully established an oligopeptide discovery strategy based on a DL model and demonstrated its utility from cytological screening to animal experimental verification.

Maternal and neonatal viromes indicate the risk of offspring's gastrointestinal tract exposure to pathogenic viruses of vaginal origin during delivery.

A cumulative effect of enterovirus and gluten intake on the risk of celiac disease autoimmunity in infants highlights the significance of viral exposure in early life on the health of children. However, pathogenic viruses may be transmitted to the offspring in an earlier period, raising the possibility that women whose vaginas are inhabited by such viruses may have had their babies infected as early as the time of delivery. A high-resolution intergenerational virome atlas was obtained by metagenomic sequencing and virome analysis on 486 samples from six body sites of 99 mother-neonate pairs. We found that neonates had less diverse oral and enteric viruses than mothers. Vaginally delivered newborns seconds after birth had a more similar oral virome and more viruses of vaginal origin than cesarean-section (C-section) newborns (56.9% vs. 5.8%). Such viruses include both Lactobacillus phage and potentially pathogenic viruses, such as herpesvirus, vaccinia virus, and hepacivirus, illustrating a relatively high variety of the pioneer viral taxa at the time of delivery and a delivery-dependent mother-to-neonate transmission along the vaginal-oral-intestinal route. Neonates are exposed to vaginal viruses as they pass through the reproductive tract, and viruses of vaginal origin may threaten their health. These findings challenge the conventional notion that vaginal delivery is definitely better than cesarean delivery from the perspective of microbial transmission. Screening for vaginal virome before delivery is a worthwhile step to advocate in normal labor to eliminate the risk of intergenerational transmission of pathogenic viruses to offspring.