Search

Abstract 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.

Cleft lip with or without palate (CL/P) is a common congenital anomaly with a high birth prevalence in China. Based on a previous linkage signal of nonsyndromic CL/P (NSCL/P) on the chromosomal region 4q31-q32 from the Chinese populations, we screened the 4q31-q32 region for susceptibility genes in 214 trios of Han Chinese. PDGF-C, an important developmental factor, resides in the region and has been implicated in NSCL/P. However, in our family-based association test (transmission disequilibrium test; TDT), we could not conclude an association between PDGF-C and NSCL/P as previously suggested. Instead, we found strong evidence for parent-of-origin effect at a PDGF-C SNP, rs17035464, by a likelihood ratio test (unadjusted p-value = 0.0018; I(m) = 2.46). The location of rs17035464 is 13 kb downstream of a previously reported, NSCL/P-associated SNP, rs28999109. Furthermore, a patient from our sample trios was observed with a maternal segmental uniparental isodisomy (UPD) in a region containing rs17035464. Our findings support the involvement of PDGF-C in the development of oral clefts; moreover, the UPD case report contributes to the collective knowledge of rare variants in the human genome.

Understanding the regional propensity differences of atherosclerosis (AS) development is hindered by the lack of animal models suitable for the study of the disease process. In this paper, we used 3SASCVD dogs, an ideal large animal human-like models for AS, to interrogate the heterogeneity of AS-prone and AS-resistant arteries; and at the single-cell level, identify the dominant cells involved in AS development. Here we present data from 3S-ASCVD dogs which reliably mimic human AS pathophysiology, predilection for lesion sites, and endpoint events. Our analysis combined bulk RNA-seq with single-cell RNA-seq to depict the transcriptomic profiles and cellular atlas of AS-prone and AS-resistant arteries in 3S-ASCVD dogs. Our results revealed the integral role of smooth muscle cells (SMCs) in regional propensity for AS. Notably, TNC+ SMCs were major contributors to AS development in 3S-ASCVD dogs, indicating enhanced extracellular matrix remodeling and transition to myofibroblasts during the AS process. Moreover, TNC+ SMCs were also present in human AS-prone carotid plaques, suggesting a potential origin of myofibroblasts and supporting the relevance of our findings. Our study provides a promising large animal model for pre-clinical studies of ASCVD and add novel insights surrounding the regional propensity of AS development in humans, which may lead to interventions that delay or prevent lesion progression and adverse clinical events. [ABSTRACT FROM AUTHOR]

Chronic Obstructive Pulmonary Disease (COPD) is a complex human disease which is driven not only by genetic factors, but also by various environmental variables, such as gender, age and smoking. Therefore, there is a demand for investigating the complexity among various risk factors involved in COPD. In this study, 44 tagging SNPs from EPHX1, GSTP1, SERPINE2 and TGFB1 were selected and genotyped in 310 COPD cases and 203 controls, all of which belong to the Han from North China. We integrated functional prediction algorithms of nonsynonymous SNPs (nsSNPs) into Bayesian network to explore the complex regulatory relationships among disease traits and various risk factors. The results showed that three basic variables (age, sex and smoking) were risk factors of COPD-related trait and phenotype. Besides these environmental risk factors, deleterious nsSNPs were found to perform better than those of significant synonymous SNPs when used as variables to make risk prediction of disease outcome. This study provides further evidences for detecting the complexity of COPD in Northern Chinese Han Population. [ABSTRACT FROM AUTHOR]

DNA variants in a 31-kb region of the human major histocompatibility complex, encompassing the tumor necrosis factor (TNF) gene cluster, were surveyed by direct sequencing of 283 unrelated individuals from six Chinese populations. A total of 273 polymorphic sites were identified, with nearly half of them novel. We observed an excess of rare variants and negative values of selection tests of the region, implying either that these populations experienced a historical expansion or that the surveyed region was subjected to natural selection. Different characteristics of the sequence variation in the six populations outline the genetic differentiation between Northern and Southern Chinese populations. The distributions of recombination rates are similar among all the populations, with variation in the magnitude and/or in the fine location of hot spots. Tag single-nucleotide polymorphisms (SNPs) selected from HapMap (Phase II) CHB data accounted for an average of 64% of common SNPs from the six Chinese populations. We also observed a limited transferability of tag SNPs between Chinese populations on the 31-kb region with an excess of untaggable SNPs and ragged linkage disequilibrium blocks. It suggested that the design and interpretation of future association studies should be more cautious, and that a resequencing approach may refine tag SNP selection on Chinese-specific disease mapping. [ABSTRACT FROM AUTHOR]