1 February 2018 in Genomics, Proteomics & Bioinformatics
Advances in biological and medical technologies have been providing us explosive volumes of biological and physiological data, such as medical images, electroencephalography, genomic and protein sequences. Learning from these data facilitates the understanding of human health and disease. Developed from artificial neural networks, deep learning-based algorithms show great promise in extracting features and learning patterns from complex data. The aim of this paper is to provide an overview of deep learning techniques and some of the state-of-the-art applications in the biomedical field. We first introduce the development of artificial neural network and deep learning. We then describe two main components of deep learning, i.e., deep learning architectures and model optimization. Subsequently, some examples are demonstrated for deep learning applications, including medical image classification, genomic sequence analysis, as well as protein structure classification and prediction. Finally, we offer our perspectives for the future directions in the field of deep learning.
26 October 2023 in Nature Machine Intelligence
Prime editors (PEs) are promising genome-editing tools, but effective optimization of prime-editing guide RNA (pegRNA) design remains a challenge owing to the lack of accurate and broadly applicable approaches. Here we develop Optimized Prime Editing Design (OPED), an interpretable nucleotide language model that leverages transfer learning to improve its accuracy and generalizability for the efficiency prediction and design optimization of pegRNAs. Comprehensive validations on various published datasets demonstrate its broad applicability in efficiency prediction across diverse scenarios. Notably, pegRNAs with high OPED scores consistently show significantly increased editing efficiencies. Furthermore, the versatility and efficacy of OPED in design optimization are confirmed by efficiently installing various ClinVar pathogenic variants using optimized pegRNAs in the PE2, PE3/PE3b and ePE editing systems. OPED consistently outperforms existing state-of-the-art approaches. We construct the OPEDVar database of optimized designs from over two billion candidates for all pathogenic variants and provide a user-friendly web application of OPED for any desired edit.
26 June 2015 in Scientific Reports
Advances in genome-wide association studies (GWAS) and large-scale sequencing studies have resulted in an impressive and growing list of disease- and trait-associated genetic variants. Most studies have emphasised the discovery of genetic variation in coding sequences, however, the noncoding regulatory effects responsible for human disease and cancer biology have been substantially understudied. To better characterise the cis-regulatory effects of noncoding variation, we performed a comprehensive analysis of the genetic variants in HOT (high-occupancy target) regions, which are considered to be one of the most intriguing findings of recent large-scale sequencing studies. We observed that GWAS variants that map to HOT regions undergo a substantial net decrease and illustrate development-specific localisation during haematopoiesis. Additionally, genetic risk variants are disproportionally enriched in HOT regions compared with LOT (low-occupancy target) regions in both disease-relevant and cancer cells. Importantly, this enrichment is biased toward disease- or cancer-specific cell types. Furthermore, we observed that cancer cells generally acquire cancer-specific HOT regions at oncogenes through diverse mechanisms of cancer pathogenesis. Collectively, our findings demonstrate the key roles of HOT regions in human disease and cancer and represent a critical step toward further understanding disease biology, diagnosis and therapy.
30 May 2024 in Nucleic Acids Research
Gene expression is temporally and spatially regulated by the interaction of transcription factors (TFs) and cis-regulatory elements (CREs). The uneven distribution of TF binding sites across the genome poses challenges in understanding how this distribution evolves to regulate spatio-temporal gene expression and consequent heritable phenotypic variation. In this study, chromatin accessibility profiles and gene expression profiles were collected from several species including mammals (human, mouse, bovine), fish (zebrafish and medaka), and chicken. Transcription factor binding sites clustered regions (TFCRs) at different embryonic stages were characterized to investigate regulatory evolution. The study revealed dynamic changes in TFCR distribution during embryonic development and species evolution. The synchronization between TFCR complexity and gene expression was assessed across species using RegulatoryScore. Additionally, an explainable machine learning model highlighted the importance of the distance between TFCR and promoter in the coordinated regulation of TFCRs on gene expression. Our results revealed the developmental and evolutionary dynamics of TFCRs during embryonic development from fish, chicken to mammals. These data provide valuable resources for exploring the relationship between transcriptional regulation and phenotypic differences during embryonic development.
10 July 2023 in Computers and Electrical Engineering