ACM-BCB ... ... : the ... ACM Conference on Bioinformatics, Computational Biology and Biomedicine. ACM Conference on Bioinformatics, Computational Biology and Biomedicine最新文献

Deep learning has shown a great promise in classifying brain disorders due to its powerful ability in learning optimal features by nonlinear transformation. However, given the high-dimension property of neuroimaging data, how to jointly exploit complementary information from multimodal neuroimaging data in deep learning is difficult. In this paper, we propose a novel multilevel convolutional neural network (CNN) fusion method that can effectively combine different types of neuroimage-derived features. Importantly, we incorporate a sequential feature selection into the CNN model to increase the feature interpretability. To evaluate our method, we classified two symptom-related brain disorders using large-sample multi-site data from 335 schizophrenia (SZ) patients and 380 autism spectrum disorder (ASD) patients within a cross-validation procedure. Brain functional networks, functional network connectivity, and brain structural morphology were employed to provide possible features. As expected, our fusion method outperformed the CNN model using only single type of features, as our method yielded higher classification accuracy (with mean accuracy >85%) and was more reliable across multiple runs in differentiating the two groups. We found that the default mode, cognitive control, and subcortical regions contributed more in their distinction. Taken together, our method provides an effective means to fuse multimodal features for the diagnosis of different psychiatric and neurological disorders.

Cryo-electron microscopy is a major structure determination technique for large molecular machines and membrane-associated complexes. Although atomic structures have been determined directly from cryo-EM density maps with high resolutions, current structure determination methods for medium resolution (5 to 10 Å) cryo-EM maps are limited by the availability of structure templates. Secondary structure traces are lines detected from a cryo-EM density map for α-helices and β-strands of a protein. When combined with secondary structure sequence segments predicted from a protein sequence, it is possible to generate a set of likely topologies of α-traces and β-sheet traces. A topology describes the overall folding relationship among secondary structures; it is a critical piece of information for deriving the corresponding atomic structure. We propose a method for protein structure prediction that combines three sources of information: the secondary structure traces detected from the cryo-EM density map, predicted secondary structure sequence segments, and amino acid contact pairs predicted using MULTICOM. A case study shows that using amino acid contact prediction from MULTICOM improves the ranking of the true topology. Our observations convey that using a small set of highly voted secondary structure contact pairs enhances the ranking in all experiments conducted for this case.

Although Cryo-electron microscopy (cryo-EM) has been successfully used to derive atomic structures for many proteins, it is still challenging to derive atomic structure when the resolution of cryo-EM density maps is in the medium range, e.g., 5-10 Å. Studies have attempted to utilize machine learning methods, especially deep neural networks to build predictive models for the detection of protein secondary structures from cryo-EM images, which ultimately helps to derive the atomic structure of proteins. However, the large variation in data quality makes it challenging to train a deep neural network with high prediction accuracy. Curriculum learning has been shown as an effective learning paradigm in machine learning. In this paper, we present a study using curriculum learning as a more effective way to utilize cryo-EM density maps with varying quality. We investigated three distinct training curricula that differ in whether/how images used for training in past are reused while the network was continually trained using new images. A total of 1,382 3-dimensional cryo-EM images were extracted from density maps of Electron Microscopy Data Bank in our study. Our results indicate learning with curriculum significantly improves the performance of the final trained network when the forgetting problem is properly addressed.

Single cell RNA sequencing is a powerful technique that measures the gene expression of individual cells in a high throughput fashion. However, due to sequencing inefficiency, the data is unreliable due to dropout events, or technical artifacts where genes erroneously appear to have zero expression. Many data imputation methods have been proposed to alleviate this issue. Yet, effective imputation can be difficult and biased because the data is sparse and high-dimensional, resulting in major distortions in downstream analyses. In this paper, we propose a completely novel approach that imputes the gene-by-gene correlations rather than the data itself. We call this method SCENA: Single cell RNA-seq Correlation completion by ENsemble learning and Auxiliary information. The SCENA gene-by-gene correlation matrix estimate is obtained by model stacking of multiple imputed correlation matrices based on known auxiliary information about gene connections. In an extensive simulation study based on real scRNA-seq data, we demonstrate that SCENA not only accurately imputes gene correlations but also outperforms existing imputation approaches in downstream analyses such as dimension reduction, cell clustering, graphical model estimation.

Image-based cell counting is a fundamental yet challenging task with wide applications in biological research. In this paper, we propose a novel Deep Network designed to universally solve this problem for various cell types. Specifically, we first extend the segmentation network, U-Net with a Self-Attention module, named SAU-Net, for cell counting. Second, we design an online version of Batch Normalization to mitigate the generalization gap caused by data augmentation in small datasets. We evaluate the proposed method on four public cell counting benchmarks - synthetic fluorescence microscopy (VGG) dataset, Modified Bone Marrow (MBM) dataset, human subcutaneous adipose tissue (ADI) dataset, and Dublin Cell Counting (DCC) dataset. Our method surpasses the current state-of-the-art performance in the three real datasets (MBM, ADI and DCC) and achieves competitive results in the synthetic dataset (VGG). The source code is available at https://github.com/mzlr/sau-net.

Protein complexes play a significant role in the core functionality of cells. These complexes are typically identified by detecting densely connected subgraphs in protein-protein interaction (PPI) networks. Recently, multiple large-scale mass spectrometry-based experiments have significantly increased the availability of PPI data in order to further expand the set of known complexes. However, high-throughput experimental data generally are incomplete, show limited agreement between experiments, and show frequent false positive interactions. There is a need for computational approaches that can address these limitations in order to improve the coverage and accuracy of human protein complexes. Here, we present a new method that integrates data from multiple heterogeneous experiments and sources in order to increase the reliability and coverage of predicted protein complexes. We first fused the heterogeneous data into a feature matrix and trained classifiers to score pairwise protein interactions. We next used graph based methods to combine pairwise interactions into predicted protein complexes. Our approach improves the accuracy and coverage of protein pairwise interactions, accurately identifies known complexes, and suggests both novel additions to known complexes and entirely new complexes. Our results suggest that integration of heterogeneous experimental data helps improve the reliability and coverage of diverse high-throughput mass-spectrometry experiments, leading to an improved global map of human protein complexes.

Next-generation sequencing (NGS) technologies offer new opportunities for precise and accurate identification of genomic aberrations, including copy number variations (CNVs). For high-throughput NGS data, using depth of coverage has become a major approach to identify CNVs, especially for whole exome sequencing (WES) data. Due to the high level of noise and biases of read-count data and complexity of the WES data, existing CNV detection tools identify many false CNV segments. Besides, NGS generates a huge amount of data, requiring to use effective and efficient methods. In this work, we propose a novel segmentation algorithm based on the total variation approach to detect CNVs more precisely and efficiently using WES data. The proposed method also filters out outlier read-counts and identifies significant change points to reduce false positives. We used real and simulated data to evaluate the performance of the proposed method and compare its performance with those of other commonly used CNV detection methods. Using simulated and real data, we show that the proposed method outperforms the existing CNV detection methods in terms of accuracy and false discovery rate and has a faster runtime compared to the circular binary segmentation method.

Stain normalization is a crucial pre-processing step for histopathological image processing, and can help improve the accuracy of downstream tasks such as segmentation and classification. To evaluate the effectiveness of stain normalization methods, various metrics based on color-perceptual similarity and stain color evaluation have been proposed. However, there still exists a huge gap between metric evaluation and human perception, given the limited explainability power of existing metrics and inability to combine color and semantic information efficiently. Inspired by the effectiveness of deep neural networks in evaluating perceptual similarity of natural images, in this paper, we propose TriNet-P, a color-perceptual similarity metric for whole slide images, based on deep metric embeddings. We evaluate the proposed approach using four publicly available breast cancer histological datasets. The benefit of our approach is its representation efficiency of the perceptual factors associated with H&E stained images with minimal human intervention. We show that our metric can capture the semantic similarities, both at subject (patient) and laboratory levels, and leads to better performance in image retrieval and clustering tasks.

Methicillin-resistant Staphylococcus aureus (MRSA) is currently the most commonly identified antibiotic-resistant pathogen in US hospitals. Resistance to methicillin is carried by SCCmec genetic elements. Multilocus sequence typing (MLST) covers internal fragments of seven housekeeping genes of S. aureus. In conjunction with mec typing, MLST has been used to create an international nomenclature for S. aureus. MLST sequence types with a single nucleotide polymorphism (SNP) considered distinct. In this work, relationships among MLST SNPs and methicillin/oxacillin resistance or susceptibility were studied, using a public data base, by means of cross-tabulation tests, multivariable (phylogenetic) logistic regression (LR), decision trees, rule bases, and random forests (RF). Model performances were assessed through multiple cross-validation. Hierarchical clustering of SNPs was also employed to analyze mutational covariation. The number of instances with a known methicillin (oxacillin) antibiogram result was 1526 (649), where 63% (54%) was resistant to methicillin (oxacillin). In univariable analysis, several MLST SNPs were found strongly associated with antibiotic resistance/susceptibility. A RF model predicted correctly the resistance/susceptibility to methicillin and oxacillin in 75% and 63% of cases (cross-validated). Results were similar for LR. Hierarchical clustering of the aforementioned SNPs yielded a high level of covariation both within the same and different genes; this suggests strong genetic linkage between SNPs of housekeeping genes and antibiotic resistant associated genes. This finding provides a basis for rapid identification of antibiotic resistant S. arues lineages using a small number of genomic markers. The number of sites could subsequently be increased moderately to increase the sensitivity and specificity of genotypic tests for resistance that do not rely on the direct detection of the resistance marker itself.

Breast cancer is a deadly disease that affects millions of women worldwide. The International Conference on Image Analysis and Recognition in 2018 presents the BreAst Cancer Histology (ICIAR2018 BACH) image data challenge that calls for computer tools to assist pathologists and doctors in the clinical diagnosis of breast cancer subtypes. Using the BACH dataset, we have developed an image classification pipeline that combines both a shallow learner (support vector machine) and a deep learner (convolutional neural network). The shallow learner and deep learners achieved moderate accuracies of 79% and 81% individually. When being integrated by fusion algorithms, the system outperformed any individual learner with the highest accuracy as 92%. The fusion presents big potential for improving clinical design support.