Search

This paper addresses the few-shot image classification problem, where the classification task is performed on unlabeled query samples given a small amount of labeled support samples only. One major challenge of the few-shot learning problem is the large variety of object visual appearances that prevents the support samples to represent that object comprehensively. This might result in a significant difference between support and query samples, therefore undermining the performance of few-shot algorithms. In this paper, we tackle the problem by proposing Few-shot Cosine Transformer (FS-CT), where the relational map between supports and queries is effectively obtained for the few-shot tasks. The FS-CT consists of two parts, a learnable prototypical embedding network to obtain categorical representations from support samples with hard cases, and a transformer encoder to effectively achieve the relational map from two different support and query samples. We introduce Cosine Attention, a more robust and stable attention module that enhances the transformer module significantly and therefore improves FS-CT performance from 5% to over 20% in accuracy compared to the default scaled dot-product mechanism. Our method performs competitive results in mini -ImageNet, CUB-200, and CIFAR-FS on 1-shot learning and 5-shot learning tasks across backbones and few-shot configurations. We also developed a custom few-shot dataset for Yoga pose recognition to demonstrate the potential of our algorithm for practical application. Our FS-CT with cosine attention is a lightweight, simple few-shot algorithm that can be applied for a wide range of applications, such as healthcare, medical, and security surveillance. The official implementation code of our Few-shot Cosine Transformer is available at https://github.com/vinuni-vishc/Few-Shot-Cosine-Transformer.

Abstract Background To date, cancer still is one of the leading causes of death worldwide, in which the cumulative of genes carrying mutations was said to be held accountable for the establishment and development of this disease mainly. From that, identification and analysis of driver genes were vital. Our previous study indicated disagreement on a unifying pipeline for these tasks and then introduced a complete one. However, this pipeline gradually manifested its weaknesses as being unfamiliar to non-technical users, time-consuming, and inconvenient. Results This study presented an R package named DrGA, developed based on our previous pipeline, to tackle the mentioned problems above. It wholly automated four widely used downstream analyses for predicted driver genes and offered additional improvements. We described the usage of the DrGA on driver genes of human breast cancer. Besides, we also gave the users another potential application of DrGA in analyzing genomic biomarkers of a complex disease in another organism. Conclusions DrGA facilitated the users with limited IT backgrounds and rapidly created consistent and reproducible results. DrGA and its applications, along with example data, were freely provided at https://github.com/huynguyen250896/DrGA .

Abstract Background When it comes to the co-expressed gene module detection, its typical challenges consist of overlap between identified modules and local co-expression in a subset of biological samples. The nature of module detection is the use of unsupervised clustering approaches and algorithms. Those methods are advanced undoubtedly, but the selection of a certain clustering method for sample- and gene-clustering tasks is separate, in which the latter task is often more complicated. Results This study presented an R-package, Overlapping CoExpressed gene Module (oCEM), armed with the decomposition methods to solve the challenges above. We also developed a novel auxiliary statistical approach to select the optimal number of principal components using a permutation procedure. We showed that oCEM outperformed state-of-the-art techniques in the ability to detect biologically relevant modules additionally. Conclusions oCEM helped non-technical users easily perform complicated statistical analyses and then gain robust results. oCEM and its applications, along with example data, were freely provided at https://github.com/huynguyen250896/oCEM .

Abstract The cumulative of genes carrying mutations is vital for the establishment and development of cancer. However, this driver gene exploring research line has selected and used types of tools and models of analysis unsystematically and discretely. Also, the previous studies may have neglected low-frequency drivers and seldom predicted subgroup specificities of identified driver genes. In this study, we presented an improved driver gene identification and analysis pipeline that comprises the four most widely focused analyses for driver genes: enrichment analysis, clinical feature association with expression profiles of identified driver genes as well as with their functional modules, and patient stratification by existing advanced computational tools integrating multi-omics data. The improved pipeline's general usability was demonstrated straightforwardly for breast cancer, validated by some independent databases. Accordingly, 31 validated driver genes, including four novel ones, were discovered. Subsequently, we detected cancer-related significantly enriched gene ontology terms and pathways, probable drug targets, two co-expressed modules associated significantly with several clinical features, such as number of positive lymph nodes, Nottingham prognostic index, and tumor stage, and two biologically distinct groups of BRCA patients. Data and source code of the case study can be downloaded at https://github.com/hauldhut/drivergene .

It has been evident that N6-methyladenosine (m6A)-modified long noncoding RNAs (m6A-lncRNAs) involves regulating tumorigenesis, invasion, and metastasis for various cancer types. In this study, we sought to pick computationally up a set of 13 hub m6A-lncRNAs in light of three state-of-the-art tools WGCNA, iWGCNA, and oCEM, and interrogated their prognostic values in brain low-grade gliomas (LGG). Of the 13 hub m6A-lncRNAs, we further detected three hub m6A-lncRNAs as independent prognostic risk factors, including HOXB-AS1, ELOA-AS1, and FLG-AS1. Then, the m6ALncSig model was built based on these three hub m6A-lncRNAs. Patients with LGG next were divided into two groups, high- and low-risk, based on the median m6ALncSig score. As predicted, the high-risk group was more significantly related to mortality. The prognostic signature of m6ALncSig was validated using internal and external cohorts. In summary, our work introduces a high-confidence prognostic prediction signature and paves the way for using m6A-lncRNAs in the signature as new targets for treatment of LGG.

Uveal melanoma (UM) is a comparatively rare cancer but requires serious consideration since patients with developing metastatic UM survive only for about 6–12 months. Fortunately, increasingly large multi-omics databases allow us to further understand cancer initiation and development. Moreover, previous studies have observed that associations between copy number aberrations (CNA) or methylation (MET) versus messenger RNA (mRNA) expression have affected these processes. From that, we decide to explore the effect of these associations on a case study of UM. Also, the current subtypes of UM display its weak association with biological phenotypes and its lack of therapy suggestions. Therefore, the re-identification of molecular subtypes is a pressing need. In this study, we recruit three omics profiles, including CNA, MET, and mRNA, in a UM cohort from The Cancer Genome Atlas (TCGA). Firstly, we identify two sets of genes, CNAexp and METexp, whose CNA and MET significantly correlated with their corresponding mRNA, respectively. Then, single and integrative analyses of the three data types are performed using the PINSPlus tool. As a result, we discover two novel integrative subgroups, IntSub1 and IntSub2, which could be a useful alternative classification for UM patients in the future. To further explore molecular events behind each subgroup, we identify their subgroup-specific genes computationally. Accordingly, the highest expressed genes among IntSub1-specific genes are mostly enriched with immune-related processes. On the other hand, IntSub2-specific genes are highly associated with cellular cation homeostasis, which responds effectively to chemotherapy using ion channel inhibitor drugs. In addition, we detect that the two integrative subgroups show different age-related risks and survival rates. These discoveries can influence the frequency of metastatic surveillance and support medical practitioners to choose an appropriate treatment regime.

Xao tam phan (Paramignya trimera (Oliv.) Guillaum) is a traditional herbal medicine in Vietnam. Previous investigations reported mainly compounds and bioactivities of roots, stems, and leaves while there is limited information about those of fruits. This study aims to reveal the difference in the chemical profile of defatted peel (DP) and nondefatted peel (NDP) methanolic extracts of P. trimera using colorimetric reactions and liquid chromatography coupled with high-resolution mass spectrometry (LC-HRMS) analysis. We also showed the potential antibacterial activity of two extracts against clinically isolated bacteria strains including P. aeruginosa, Salmonella sp., and S. aureus with the MIC values

The unprecedented proliferation of recent large-scale and multi-omics databases of cancers has given us many new insights into genomic and epigenomic deregulation in cancer discovery in general. However, we wonder whether or not there exists a systematic connection between copy number aberrations (CNA) and methylation (MET)? If so, what is the role of this connection in breast cancer (BRCA) tumorigenesis and progression? At the same time, the PAM50 intrinsic subtypes of BRCA have gained the most attention from BRCA experts. However, this classification system manifests its weaknesses including low accuracy as well as a possible lack of association with biological phenotypes, and even further investigations on their clinical utility were still needed. In this study, we performed an integrative analysis of three-omics profiles, CNA, MET, and mRNA expression, in two BRCA patient cohorts (one for discovery and another for validation) – to elucidate those complicated relationships. To this purpose, we first established a set of CNAcor and METcor genes, which had CNA and MET levels significantly correlated (and anti-correlated) with their corresponding expression levels, respectively. Next, to revisit the current classification of BRCA, we performed single and integrated clustering analyses using our clustering method PINSPlus. We then discovered two biologically distinct subgroups that could be an improved and refined classification system for breast cancer patients, which can be validated by a third-party data. Further studies were then performed and realized each-subgroup-specific genes and different interactions between each of the two identified subgroups with the age factor. These findings can show promise as diagnostic and prognostic values in BRCA, and a potential alternative to the PAM50 intrinsic subtypes in the future.

This paper presents an experimental study on the seismic performance of a novel composite joint between reinforced concrete columns and steel beams (RCSs) for frame structures. In the proposed RCS joint, an H steel profile totally embedded inside an RC column is directly welded to the steel beam. The H steel profile was covered by two supplementary plates to avoid the stirrups resisting the shear in the connection region. Two full-scale joints were built and tested under reversed-cyclic loading at the University of Transport and Communications of Vietnam. Seismic performance in terms of load-bearing capacity, story drift capacity, energy dissipation, and stiffness degradation were evaluated. The test specimen showed a good response to cyclic load reversals.

The interface bond between layers plays an important role in the behavior of pavement structure. However, this aspect has not yet been adequately considered in the pavement analysis process due to the lack of advanced characterizations of actual condition. In many pavement design procedures, only completely bonded or unbounded interfaces between the layers are considered. For the purpose of the better evaluation of the asphalt pavement behavior, this work focused on its investigation taking into account the actual interface bonding condition between the asphalt layers. Based on the layered theory developed by Burmister (1943), the actual interaction between pavement layers was taken into account by introducing a horizontal shear reaction modulus which represents the interface bonding condition for a given state. The analytical solution was then implemented in a numerical program before doing forward calculations for sensitivity analysis which highlights the influence of the interface bonding conditions on the structural behaviors of asphalt pavement under a static load. Furthermore, the numerical program was applied through an original experimental case study where falling weight deflectometer (FWD) tests were carried out on two full-scale pavement structures with or without a geogrid at the interface between the asphalt layers. Backcalculations of the FWD measurements allowed determining field condition of the interface bond between the asphalt layers. The obtained values of the interface shear modulus in pavement structure with a geogrid are smaller than the ones in pavement structure without geogrid. Moreover, all of these values representing field performance are at the same order of magnitude as those from dynamic interlayer shear testing.

Abstract Glass biodeterioration by fungi has caused irreversible damage to valuable glass materials such as cultural heritages and optical devices. To date, knowledge about metabolic potential and genomic profile of biodeteriorative fungi is still scarce. Here, we report for the first time the whole genome sequence of Curvularia eragrostidis C52 that strongly degraded silica-based glasses coated with fluorine and hafnium, as expressed by the hyphal surface coverage of 46.16 ± 3.3% and reduced light transmission of 50.93 ± 1.45%. The genome of C. eragrostidis C52 is 36.9 Mb long with a GC content of 52.1% and contains 14,913 protein-coding genes, which is the largest genome ever recorded in the genus Curvularia. Phylogenomic analysis revealed C. eragrostidis C52 formed a distinct cluster with Curvularia sp. IFB-Z10 and was not evolved from compared genomes. Genome-wide comparison showed that strain C52 harbored significantly higher proportion of proteins involved in carbohydrate-active enzymes, peptidases, secreted proteins, and transcriptional factors, which may be potentially attributed to a lifestyle adaptation. Furthermore, 72 genes involved in the biosynthesis of 6 different organic acids were identified and expected to be crucial for the fungal survival in the glass environment. To form biofilm against stress, the fungal strain utilized 32 genes responsible for exopolysaccharide production. These findings will foster a better understanding of the biology of C. eragrostidis and the mechanisms behind fungal biodeterioration in the future.