Biology subjects

Sato, K.

Publications and source records attributed to Sato, K..

173 records · Page 10Linked to original sources

bioRxiv2019

Quantifying normal and parkinsonian gait features from home movies: Practical application of a deep learning-based 2D pose estimator

ObjectiveGait movies recorded in daily clinical practice are usually not filmed with specific devices, which prevents neurologists benefitting from leveraging gait analysis technologies. Here we propose a novel unsupervised approach to quantifying gait features and to extract cadence from normal and parkinsonian gait movies recorded with a home video camera by applying OpenPose, a deep learning-based 2D-pose estimator that can obtain joint coordinates from pictures or videos recorded with a monocular camera.\n\nMethodsOur proposed method consisted of two distinct phases: obtaining sequential gait features from movies by extracting body joint coordinates with OpenPose; and estimating cadence of periodic gait steps from the sequential gait features using the short-time pitch detection approach.\n\nResultsThe cadence estimation of gait in its coronal plane (frontally viewed gait) as is frequently filmed in the daily clinical setting was successfully conducted in normal gait movies using the short-time autocorrelation function (ST-ACF). In cases of parkinsonian gait with prominent freezing of gait and involuntary oscillations, using ACF-based statistical distance metrics, we quantified the periodicity of each gait sequence; this metric clearly corresponded with the subjects baseline disease statuses.\n\nConclusionThe proposed method allows us to analyze gait movies that have been underutilized to date in a completely data-driven manner, and might broaden the range of movies for which gait analyses can be conducted.

neuroscience↗

bioRxiv2019

Rapid evolution of antiviral APOBEC3 genes driven by the conflicts with ancient retroviruses

The evolution of antiviral genes has been fundamentally shaped by antagonistic interactions with ancestral viruses. The AID/APOBEC family genes (AID and APOBEC1-4) encode cellular cytosine deaminases that target nucleic acids and catalyze C-to-U mutations. In the case of retroviral replication, APOBEC3 proteins induce C-to-U mutations in minus-stranded viral DNA, which results in G-to-A mutations in the viral genome. Previous studies have indicated that the expansion and rapid evolution of mammalian APOBEC3 genes has been driven by an arms race with retroviral parasites, but this has not been thoroughly investigated. Endogenous retroviruses (ERVs) are retrotransposons originated from ancient retroviral infections. These sequences sometimes bear the hallmarks of APOBEC3-mediated mutations, and therefore serve as a record of the ancient conflict between retroviruses and APOBEC3 genes. Here we systematically investigated the sequences of ERVs and APOBEC3 genes in mammals to reconstruct details of the evolutionary conflict between them. We identified 1,420 AID/APOBEC family genes in a comprehensive screen of mammalian genome. Of the AID/APOBEC family genes, APOBEC3 genes have been selectively amplified in mammalian genomes and disclose evidence of strong positive selection - whereas the catalytic domain was highly conserved across species, the structure loop 7, which recognizes viral DNA/RNA substrates, was shown to be evolving under strong positive selection. Although APOBEC3 genes have been amplified by tandem gene duplication in most mammalian lineages, the retrotransposition-mediated gene amplification was found in several mammals including New World monkeys and prosimian primates. Comparative analysis revealed that G-to-A mutations are accumulated in ERVs, and that the G-to-A mutation signatures on ERVs is concordant with the target preferences of APOBEC3 proteins. Importantly, the number of APOBEC3 genes was significantly correlated with the frequency of G-to-A mutations in ERVs, suggesting that the amplification of APOBEC3 genes led to stronger attacks on ERVs and/or their ancestral retroviruses by APOBEC3 proteins. Furthermore, the numbers of APOBEC3 genes and ERVs in mammalian genomes were positively correlated, and in primates, the timings of APOBEC3 gene amplification was concordant with that of ERV invasions. Our findings suggest that conflict with ancient retroviruses was a major selective pressure driving the rapid evolution of APOBEC3 genes in mammals.

evolutionary biology↗

bioRxiv2019

The whale shark genome reveals patterns of vertebrate gene family evolution

Chondrichthyes (cartilaginous fishes) are fundamental for understanding vertebrate evolution, yet their genomes are understudied. We report long-read sequencing of the whale shark genome to generate the best gapless chondrichthyan genome assembly yet with higher contig contiguity than all other cartilaginous fish genomes, and studied vertebrate genomic evolution of ancestral gene families, immunity, and gigantism. We found a major increase in gene families at the origin of gnathostomes (jawed vertebrates) independent of their genome duplication. We studied vertebrate pathogen recognition receptors (PRRs), which are key in initiating innate immune defense, and found diverse patterns of gene family evolution, demonstrating that adaptive immunity in gnathostomes did not fully displace germline-encoded PRR innovation. We also discovered a new Toll-like receptor (TLR29) and three NOD1 copies in the whale shark. We found chondrichthyan and giant vertebrate genomes had decreased substitution rates compared to other vertebrates, but gene family expansion rates varied among vertebrate giants, suggesting substitution and expansion rates of gene families are decoupled in vertebrate genomes. Finally, we found gene families that shifted in expansion rate in vertebrate giants were enriched for human cancer-related genes, consistent with gigantism requiring adaptations to suppress cancer.

genomics↗

bioRxiv2019

Benchmarking principal component analysis for large-scale single-cell RNA-sequencing

Principal component analysis (PCA) is an essential method for analyzing single-cell RNA-seq (scRNA-seq) datasets, but large-scale scRNA-seq datasets require long computational times and a large memory capacity.\n\nIn this work, we review 21 fast and memory-efficient PCA implementations (10 algorithms) and evaluate their application using 4 real and 18 synthetic datasets. Our benchmarking showed that some PCA algorithms are faster, more memory efficient, and more accurate than others. In consideration of the differences in the computational environments of users and developers, we have also developed guidelines to assist with selection of appropriate PCA implementations.

bioinformatics↗

bioRxiv2018

cTAGE5 acts as a Sar1 GTPase regulator for collagen export

Secretory proteins synthesized within the endoplasmic reticulum (ER) are exported via coat protein complex II (COPII)-coated vesicles. The formation of the COPII-coated vesicles is initiated by activation of the small GTPase, Sar1. cTAGE5 directly interacts with a guanine-nucleotide exchange factor (GEF), Sec12, and a GTPase-activating protein (GAP) of Sar1, Sec23. We have previously shown that cTAGE5 recruits Sec12 to the ER exit sites for efficient production of activated Sar1 for collagen secretion. However, the functional significance of the interaction between cTAGE5 and Sec23 has not been fully elucidated. In this study, we showed that cTAGE5 enhances the GAP activity of Sec23 toward Sar1. In addition, the interaction of cTAGE5 with Sec23 is necessary for collagen exit from the ER. Our data suggests that cTAGE5 acts as a Sar1 GTPase regulator for collagen secretion.

cell biology↗

bioRxiv2018

HSF2BP Negatively Regulates Homologous Recombination in DNA Interstrand Crosslink Repair in Human Cells by Direct Interaction With BRCA2

The tumor suppressor BRCA2 is essential for homologous recombination, replication fork stability and DNA interstrand crosslink (ICL) repair in vertebrates. We show that a functionally uncharacterized protein, HSF2BP, is involved in a novel, direct and highly evolutionarily conserved interaction with BRCA2. Although HSF2BP was previously described as testis-specific, we find it is expressed in mouse ES cells, in human cancer cell lines, and in tumor samples. Elevated levels of HSF2BP sensitize human cells to ICL-inducing agents (mitomycin C and cisplatin) and PARP inhibitors, resulting in a phenotype characteristic of cells from Fanconi anemia (FA) patients. We biochemically recapitulate the suppression of ICL repair and establish that excess HSF2BP specifically compromises homologous recombination by preventing BRCA2 and RAD51 loading at the ICL. As increased ectopic expression of HSF2BP occurs naturally, we suggest that it can be considered as a causative agent in FA and a source of cancer-promoting genomic instability.

molecular biology↗

bioRxiv2018

CellFishing.jl: an ultrafast and scalable cell search method for single-cell RNA-sequencing

Recent technical improvements in single-cell RNA sequencing (scRNA-seq) have enabled massively parallel profiling of transcriptomes, thereby promoting large-scale studies encompassing a wide range of cell types of multicellular organisms. With this background, we propose CellFishing.jl, a new method for searching atlas-scale datasets for similar cells and detecting noteworthy genes of query cells with high accuracy and throughput. Using multiple scRNA-seq datasets, we validate that our method demonstrates comparable accuracy to and is markedly faster than the state-of-the-art software. Moreover, CellFishing.jl is scalable to more than one million cells, and the throughput of the search is approximately 1,600 cells per second.

bioinformatics↗

bioRxiv2018

Gorilla APOBEC3 restricts SIVcpz and influences lentiviral evolution in great ape cross-species transmissions

Restriction factors including APOBEC3 family proteins have the potential prevent cross-species lentivirus transmissions. Such events as well as ensuing pathogenesis require the viral Vif protein to overcome/neutralize/degrade the APOBEC3 enzymes of the new host species. Previous investigations have focused on the molecular interaction between human APOBEC3s and HIV-1 Vif. However, the evolutionary interplay between lentiviruses and great ape (including human, chimpanzee and gorilla) APOBEC3s has not been fully investigated. Here we demonstrate that gorilla APOBEC3G plays a pivotal role in restricting lentiviral transmission from chimpanzee to gorilla. We also reveal that a sole amino acid substitution in Vif is sufficient to overcome the gorilla APOBEC3G-mediated species barrier. Moreover, the antiviral effects of gorilla APOBEC3D and APOBEC3F are considerably weaker than those of human and chimpanzee counterparts, which can result in the skewed evolution of great ape lentiviruses leading to HIV-1.\n\nHighlightsO_LISIVcpz requires M16E mutation in Vif to counteract gorilla A3G\nC_LIO_LIAcidic residue at position 16 of Vif is crucial to counteract gorilla A3G\nC_LIO_LIGorilla A3D and A3F poorly suppress lentiviral infectivity\nC_LIO_LISIVgor and related HIV-1s counteract human A3D and A3F independently of DRMR motif\nC_LI

microbiology↗

bioRxiv2018

A new targeted capture method using bacterial artificial chromosome (BAC) libraries as baits for sequencing relatively large genes

To analyze a specific genome region using next-generation sequencing technologies, the enrichment of DNA libraries with targeted capture methods has been standardized. For enrichment of mitochondrial genome, a previous study developed an original targeted capture method that use baits constructed from long-range polymerase chain reaction (PCR) amplicons, common laboratory reagents, and equipment. In this study, a new targeted capture method is presented, that of bacterial artificial chromosome (BAC) double capture (BDC), modifying the previous method, but using BAC libraries as baits for sequencing a relatively large gene. We applied the BDC approach for the 214 kb autosomal region, ring finger protein 213, which is the susceptibility gene of moyamoya disease (MMD). To evaluate the reliability of BDC, cost and data quality were compared with those of a commercial kit. While the ratio of duplicate reads was higher, the cost was less than that of the commercial kit. The data quality was sufficiently the same as that of the kit. Thus, BDC can be an easy, low-cost, and useful method for analyzing individual genome region with substantial length.

genomics↗

bioRxiv2018

Direct inference of base-pairing probabilities with neural networks improves RNA secondary structure prediction with pseudoknots

MotivationExisting approaches for predicting RNA secondary structures depend on howto decompose a secondary structure into substructures, so-called the architecture, to define their parameter space. However, the architecture has not been sufficiently investigated especially for pseudoknotted secondary structures.\n\nResultsIn this paper, we propose a novel algorithm to directly infer base-pairing probabilities with neural networks that does not depend on the architecture of RNA secondary structures, followed by performing the maximum expected accuracy (MEA) based decoding algorithms; Nussinov-style decoding for pseudoknot-free structures, and IPknot-style decoding for pseudoknotted structures. To train the neural networks connected to each base-pair, we adopt a max-margin framework, called structured support vector machines (SSVM), as the output layer. Our benchmarks for predicting RNA secondary structures with and without pseudoknots show that our algorithm achieves the best prediction accuracy compared with existing methods.\n\nAvailabilityThe source code is available at https://github.com/keio-bioinformatics/neuralfold/.\n\nContactsatoken@bio.keio.ac.jp

bioinformatics↗

bioRxiv2017

A max-margin training of RNA secondary structure prediction integrated with the thermodynamic model

Motivation: A popular approach for predicting RNA secondary structure is the thermodynamic nearest neighbor model that finds a thermodynamically most stable secondary structure with the minimum free energy (MFE). For further improvement, an alternative approach that is based on machine learning techniques has been developed. The machine learning based approach can employ a fine-grained model that includes much richer feature representations with the ability to fit the training data. Although a machine learning based fine-grained model achieved extremely high performance in prediction accuracy, a possibility of the risk of overfitting for such model has been reported.\n\nResults: In this paper, we propose a novel algorithm for RNA secondary structure prediction that integrates the thermodynamic approach and the machine learning based weighted approach. Ourfine-grained model combines the experimentally determined thermodynamic parameters with a large number of scoring parameters for detailed contexts of features that are trained by the structured support vector machine (SSVM) with the{ell} 1 regularization to avoid overfitting. Our benchmark shows that our algorithm achieves the best prediction accuracy compared with existing methods, and heavy overfitting cannot be observed.\n\nAvailability: The implementation of our algorithm is available at https://github.com/keio-bioinformatics/mxfold.\n\nContact: satoken@bio.keio.ac.jp

bioinformatics↗