Distal vaginal atresia is a rare abnormality of female reproductive tract in which the vagina is closed or absent. The distal vagina may be replaced by fibrous tissue and the condition is often not diagnosed until a girl fails to begin having periods at puberty. Although it is a congenital disorder, potential genetic causes of distal vaginal atresia are still unknown. We recruited a cohort of 39 patients with distal vaginal atresia and analyzed their phenotypic and genetic features. In addition to the complaint of distal vaginal atresia, approximately 17.9% (7/39) of the patients had other Müllerian anomalies, and 17.9% (7/39) of the patients had other structural abnormalities, including renal-tract, skeletal and cardiac anomalies. Using genome sequencing, we identified two fragment duplications on 17q12 encompassing HNF1B and LHX1, two dosage-sensitive genes with candidate pathogenic variants, in two unrelated patients. A large fragment of uniparental disomy was detected in another patient, affecting genes involved in cell morphogenesis and connective tissue development. Additionally, we reported two variants on TBX3 and AXL, leading to distal vaginal atresia in mutated mouse model, in our clinical subjects for the first time. Essential biological functions of these detected genes with pathogenic variants included regulating reproductive development and cell fate and patterning during embryogenesis. We displayed the comprehensive clinical and genetic characteristic of a cohort with distal vaginal atresia and they were highly heterogeneous both phenotypically and genetically. The duplication of 17q12 in our cohort could help to expand its phenotypic spectrum and potential contribution to the distal vaginal atresia. Our findings of pathogenic genetic variants and associated phenotypes in our cohort could provide evidence and new insight for further research attempting to reveal genetic causes of distal vaginal atresia.

Cell-free RNA, including both long RNA and small RNA, has been considered important for its biological functions and potential clinical usage, but the major challenge is to effectively sequence them at the same time. Here we present PolyAdenylation Ligation Mediated-Seq (PALM-Seq), an integrated sequencing method for cell-free long and small RNA. Through terminal modification and addition of 3’ polyadenylation and 5’ adaptor, we could get mRNA, long non-coding RNA, microRNA, tRNA, piRNA and other RNAs in a single library. With target RNA depletion, all these RNAs could be sequenced with relatively low depth. Using PALM-Seq, we identified pregnant-related mRNAs, long non-coding RNAs and microRNAs in female plasma. We also applied PALM-Seq to sequence RNA from amniotic fluids, leukocytes and placentas, and could find RNA signatures associated with specific sample type. PALM-Seq provides an integrated, cost-effective and simple method to characterize the landscape of cell-free RNA, and can stimulate further progress in cell-free RNA study and usage.

Background: Copy-number variants (CNVs) have been recognized as one of the major causes of genetic disorders. Reliable detection of CNVs from genome sequencing data has been a strong demand for disease research. However, current software for detecting CNVs has high false-positive rates, which needs further improvement. Methods: Here, we proposed a novel and post-processing approach for CNVs prediction (CNV-P), a machine-learning framework that could efficiently remove false-positive fragments from results of CNVs detecting tools. A series of CNVs signals such as read depth (RD), split reads (SR) and read pair (RP) around the putative CNV fragments were defined as features to train a classifier. Results: The prediction results on several real biological datasets showed that our models could accurately classify the CNVs at over 90% precision rate and 85% recall rate, which greatly improves the performance of state-of-the-art algorithms. Furthermore, our results indicate that CNV-P is robust to different sizes of CNVs and the platforms of sequencing. Conclusions: Our framework for classifying high-confident CNVs could improve both basic research and clinical diagnosis of genetic diseases.

Analyzing cell types of origin of cell-free RNA can enhance the resolution of liquid biopsies, thereby deepening the understanding of molecular and cellular changes in development and disease processes. Existing deconvolution methods typically rely on meticulously curated gene expression profiles or employ deep neural network with vast and complex solution spaces that are difficult to interpret. These approaches overlook the synergistic and co-expression effects among genes in biological signaling pathways, compromising their generalizability and robustness. we developed ‘Deconformer’, a Transformer-based deconvolution model that integrates biological signaling pathways at the embedding stage, to address these issues. Compared to popular methods on multiple datasets, Deconformer demonstrates superior performance and robustness, and is capable of tracking the developmental process of the fetal and placenta. Additionally, pathway-level interpretability of Deconformer offers new insights into crosstalk, dependencies, and other interactions within cell-free RNA pathways, supporting further biological discoveries. We posit that Deconformer represents a significant advancement in the precise analysis of the cell-free transcriptome. It holds the promise of describing disease progression and severity with a new level of accuracy, focusing on the contributions of originating cell types and their pathway dependencies. This model has the potential to catalyze the development of non-invasive diagnostic tools and enhance our understanding of the underlying biology of diseases.