Holdtsvensson0425

Automatic segmentation and localization of lesions in mammogram (MG) images are challenging even with employing advanced methods such as deep learning (DL) methods. We developed a new model based on the architecture of the semantic segmentation U-Net model to precisely segment mass lesions in MG images. The proposed end-to-end convolutional neural network (CNN) based model extracts contextual information by combining low-level and high-level features. We trained the proposed model using huge publicly available databases, (CBIS-DDSM, BCDR-01, and INbreast), and a private database from the University of Connecticut Health Center (UCHC).

We compared the performance of the proposed model with those of the state-of-the-art DL models including the fully convolutional network (FCN), SegNet, Dilated-Net, original U-Net, and Faster R-CNN models and the conventional region growing (RG) method. The proposed Vanilla U-Net model outperforms the Faster R-CNN model significantly in terms of the runtime and the Intersectn, and modifying the architecture of the original model results in better performance in terms of the mean accuracy, the mean DI, and the mean IOU in detecting mass lesion compared to the other DL and the conventional models.

The proposed Vanilla U-Net based model can be used for precise segmentation of masses in MG images. This is because the segmentation process incorporates more multi-scale spatial context, and captures more local and global context to predict a precise pixel-wise segmentation map of an input full MG image. These detected maps can help radiologists in differentiating benign and malignant lesions depend on the lesion shapes. We show that using transfer learning, introducing augmentation, and modifying the architecture of the original model results in better performance in terms of the mean accuracy, the mean DI, and the mean IOU in detecting mass lesion compared to the other DL and the conventional models.

Annually, over 1 billion people sustain traumatic injuries, resulting in over 900,000 deaths in Africa and 6 million deaths globally. Timely response, intervention, and transportation in the prehospital setting reduce morbidity and mortality of trauma victims. Our objective was to describe the existing literature evaluating trauma morbidity and mortality outcomes as a function of prehospital care time to identify gaps in literature and inform future investigation.

We performed a scoping review of published literature in MEDLINE. Results were limited to English language publications from 2009 to 2020. Included articles reported trauma outcomes and prehospital time. We excluded case reports, reviews, systematic reviews, meta-analyses, comments, editorials, letters, and conference proceedings. In total, 808 articles were identified for title and abstract review. Of those, 96 articles met all inclusion criteria and were fully reviewed. Higher quality studies used data derived from trauma registries. There wasations in Africa and LMICs.

The existing literature disproportionately represents high-income settings and most commonly assessed in-hospital mortality as a function of crude prehospital time. Future studies should focus on how specific prehospital intervals impact morbidity outcomes (e.g., organ failure) and mortality at earlier time points (e.g., 3 or 7 days) to better reflect the effect of early prehospital resuscitation and transport. Trauma registries may be a tool to facilitate such research and may promote higher quality investigations in Africa and LMICs.

Monoculture farming poses significant disease challenges, but fungus-farming termites are able to successfully keep their monoculture crop free from contamination by other fungi. It has been hypothesised that obligate gut passage of all plant substrate used to manure the fungal symbiont is key to accomplish this. see more Here we refute this hypothesis in the fungus-farming termite species Macrotermes bellicosus.

We first used ITS amplicon sequencing to show that plant substrate foraged on by termite workers harbour diverse fungal communities, which potentially could challenge the farming symbiosis. Subsequently, we cultivated fungi from dissected sections of termite guts to show that fungal diversity does not decrease during gut passage. Therefore, we investigated if healthy combs harboured these undesirable fungal genera, and whether the presence of workers affected fungal diversity within combs. Removal of workers led to a surge in fungal diversity in combs, implying that termite defences must be responsible fove insurmountable.

Identifying one or more biologically-active/native decoys from millions of non-native decoys is one of the major challenges in computational structural biology. The extreme lack of balance in positive and negative samples (native and non-native decoys) in a decoy set makes the problem even more complicated. Consensus methods show varied success in handling the challenge of decoy selection despite some issues associated with clustering large decoy sets and decoy sets that do not show much structural similarity. Recent investigations into energy landscape-based decoy selection approaches show promises. However, lack of generalization over varied test cases remains a bottleneck for these methods.

We propose a novel decoy selection method, ML-Select, a machine learning framework that exploits the energy landscape associated with the structure space probed through a template-free decoy generation. The proposed method outperforms both clustering and energy ranking-based methods, all the while consistently offering better performance on varied test-cases. Moreover, ML-Select shows promising results even for the decoy sets consisting of mostly low-quality decoys.

ML-Select is a useful method for decoy selection. This work suggests further research in finding more effective ways to adopt machine learning frameworks in achieving robust performance for decoy selection in template-free protein structure prediction.

ML-Select is a useful method for decoy selection. This work suggests further research in finding more effective ways to adopt machine learning frameworks in achieving robust performance for decoy selection in template-free protein structure prediction.

The RNA interference (RNAi) pathway is a gene regulation mechanism that utilizes small RNA (sRNA) and Argonaute (Ago) proteins to silence target genes. Our previous work identified a functional RNAi pathway in the protozoan parasite Entamoeba histolytica, including abundant 27 nt antisense sRNA populations which associate with EhAgo2-2 protein. However, there is lack of understanding about the sRNAs that are bound to two other EhAgos (EhAgo2-1 and 2-3), and the mechanism of sRNA regulation itself is unclear in this parasite. Therefore, identification of the entire pool of sRNA species and their sub-populations that associate with each individual EhAgo protein would be a major step forward.

In the present study, we sequenced sRNA libraries from both total RNAs and EhAgo bound RNAs. We identified a new population of 31 nt sRNAs that results from the addition of a non-templated 3-4 adenosine nucleotides at the 3'-end of the 27 nt sRNAs, indicating a non-templated RNA-tailing event in the parasite. The relatiodification, which is the first such observation amongst single celled protozoan parasites. Our sRNA sequencing libraries provide the first comprehensive sRNA dataset for all three Entamoeba Ago proteins, which can serve as a useful database for the amoeba community.

We identified a new population of sRNA with non-templated oligo-adenylation modification, which is the first such observation amongst single celled protozoan parasites. Our sRNA sequencing libraries provide the first comprehensive sRNA dataset for all three Entamoeba Ago proteins, which can serve as a useful database for the amoeba community.

Drug repositioning has been an important and efficient method for discovering new uses of known drugs. Researchers have been limited to one certain type of collaborative filtering (CF) models for drug repositioning, like the neighborhood based approaches which are good at mining the local information contained in few strong drug-disease associations, or the latent factor based models which are effectively capture the global information shared by a majority of drug-disease associations. Few researchers have combined these two types of CF models to derive a hybrid model which can offer the advantages of both. Besides, the cold start problem has always been a major challenge in the field of computational drug repositioning, which restricts the inference ability of relevant models.

Inspired by the memory network, we propose the hybrid attentional memory network (HAMN) model, a deep architecture combining two classes of CF models in a nonlinear manner. First, the memory unit and the attention mechanism are comgs.

Through the performance on two drug repositioning data sets, we believe that the HAMN model proposes a new solution to improve the prediction accuracy of drug-disease associations and give pharmaceutical personnel a new perspective to develop new drugs.

RNA-binding proteins (RBPs) play crucial roles in various biological processes. Deep learning-based methods have been demonstrated powerful on predicting RBP sites on RNAs. However, the training of deep learning models is very time-intensive and computationally intensive.

Here we present a deep learning-based RBPsuite, an easy-to-use webserver for predicting RBP binding sites on linear and circular RNAs. For linear RNAs, RBPsuite predicts the RBP binding scores with them using our updated iDeepS. For circular RNAs (circRNAs), RBPsuite predicts the RBP binding scores with them using our developed CRIP. RBPsuite first breaks the input RNA sequence into segments of 101 nucleotides and scores the interaction between the segments and the RBPs. RBPsuite further detects the verified motifs on the binding segments gives the binding scores distribution along the full-length sequence.

RBPsuite is an easy-to-use online webserver for predicting RBP binding sites and freely available at http//www.csbio.sjtu.edu.cn/bioinf/RBPsuite/ .

RBPsuite is an easy-to-use online webserver for predicting RBP binding sites and freely available at http//www.csbio.sjtu.edu.cn/bioinf/RBPsuite/ .Generally, autoimmune diseases are more prevalent in females than males. Various predisposing factors, including female sex hormones, X chromosome genes, and the microbiome have been implicated in the female bias of autoimmune diseases. During embryogenesis, one of the X chromosomes in the females is transcriptionally inactivated, in a process called X chromosome inactivation (XCI). This equalizes the impact of two X chromosomes in the females. However, some genes escape from XCI, providing a basis for the dual expression dosage of the given gene in the females. In the present review, the contribution of the escape genes to the female bias of autoimmune diseases will be discussed.

Late blight disease (LBD) caused by the pathogen Phytophthora infestans (PI), is the most devastating disease limiting potato (Solanum tuberosum) production globally. Currently, this disease pathogen is re-emerging and appearing in new areas at a very high intensity. A better understanding of the natural defense mechanisms against PI in different potato cultivars especially at the protein level is still lacking. Therefore, to elucidate potato proteome response to PI, we investigated changes in the proteome and leaf morphology of three potato cultivars, namely; Favorita (FA), Mira (MA), and E-malingshu N0.14 (E14) infected with PI by using the iTRAQ-based quantitative proteomics analysis.

A total of 3306 proteins were found in the three potato genotypes, and 2044 proteins were quantified. Cluster analysis revealed MA and E14 clustered together separately from FA. The protein profile and related functions revealed that the cultivars shared a typical hypersensitive response to PI, including induction of elicitors, oxidative burst, and suppression of photosynthesis in the potato leaves.