Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics最新文献

{"title":"MinIsoClust","authors":"S. Behera, J. Deogun, E. Moriyama","doi":"10.1145/3388440.3412424","DOIUrl":"https://doi.org/10.1145/3388440.3412424","url":null,"abstract":"With the advent of next-generation sequencing technologies, computational transcriptome assembly of RNA-Seq data has become a critical step in many biological and biomedical studies. The accuracy of these transcriptome assembly methods is hindered by the presence of alternatively spliced transcripts (isoforms). Identifying and quantifying isoforms is also essential in understanding complex biological functions, many of which are often associated with various diseases. However, clustering of isoform sequences using only sequence identities when quality reference genomes are not available is often difficult due to heterogeneous exon composition among isoforms. Clustering of a large number of transcript sequences also requires a scalable technique. In this study, we propose a minwise-hashing based method, MinIsoClust, for fast and accurate clustering of transcript sequences that can be used to identify groups of isoforms. We tested this new method using simulated datasets. The results demonstrated that MinIso-Clust was more accurate than CD-HIT-EST, isONclust, and MM-seqs2/Linclust. MinIsoClust also performed better than isONclust and MMseqs2/Linclust in terms of computational time and space efficiency. The source codes of MinIsoClust is freely available at https://github.com/srbehera/MinIsoClust.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125319715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multi-context feature learning approach to identify disease-specific gene neighborhoods","authors":"S. Ghandikota, A. Jegga","doi":"10.1145/3388440.3412419","DOIUrl":"https://doi.org/10.1145/3388440.3412419","url":null,"abstract":"Analyzing gene networks in a specific phenotype state can provide important insights into pathways and biological processes underlying the onset and progression of the disease. Specifically, analyzing gene neighborhoods around key disease-driver genes and transcription factors can lead to discovery of regulatory networks and novel therapeutic targets. Traditional methods to decipher these regulatory networks mostly rely on transcriptomic signals and do not incorporate the different functional contexts available, making them inadequate to model the inherently complex relationships between genes and their neighborhoods. We present a neural network-based representation learning framework which uses both co-expression and functional gene contexts to learn continuous gene representations. It can be used to extract distributed representations of genes in normal (e.g., control, wild-type, etc.) and perturbed states (e.g., disease, knockout, etc.) by integrating co-expressed gene pairs from multiple transcriptomic datasets. To show the utility of this approach, we trained our model on whole lung tissue transcriptomic studies of idiopathic pulmonary fibrosis (IPF) to generate disease-specific gene representations. We compare the gene features from our method with two other representation learning methods by generating and analyzing the regulatory gene neighborhoods of known transcription factors in the lung tissue. Using several TF-target gene set libraries, we show that the regulatory gene neighborhoods by our method are biologically relevant.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125438715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparing Type 2 Diabetes Self-Management Apps Against the Needs of Low-Income Minority Patients: Is There An Implicit Functionality Bias?","authors":"Wayne W. Zachary, Hita Gupta","doi":"10.1145/3388440.3414913","DOIUrl":"https://doi.org/10.1145/3388440.3414913","url":null,"abstract":"Background: Diabetes Mellitus is a chronic disease affecting 30 million in the US. It is a leading cause of death and a major risk factor for severe COVID-19. More than 90% of cases are Type 2 (T2DM), which has adult onset and has risk factors that are behavioral (e.g., smoking) or environmental (e.g., poor nutrition, decreased physical activity). Self-management is critical to long-term treatment of T2DM. It includes adherence to medication regimens, constant nutritional and physical activity management, blood glucose monitoring, and behavioral changes (e.g., smoking cessation). Many mobile computing health (mHealth) apps have been developed to support TM self-management. Problem: US T2DM rates among non-Hispanic whites and the well-educated have leveled off, but diagnoses continue to increase disproportionately among low-income populations, particularly African-American, Latino, and Native American minorities. This has created a growing health disparity associated with social and economic factors that include differential access to healthcare, healthy food, occupational opportunities and physical activity options. (termed Social Determinants of Health or SDOH [1]. Recent public health research [2,3] has begun to identify unique SDOH challenges faced by one such population, low-income African Americans. This poster examines the degree to which the existing T2DM mHealth apps are able to address the self-management needs exposed in this emerging research, versus the more widely studied needs and issues associated with more affluent and largely white population of persons with T2DM. Methods: Seventeen positively assessed T2DM apps were selected from recent review articles. Separately, two sets of functional features were compiled. First, from the T2DM literature, a set of 23 categories and sub-categories was compiled of general features that were identified as desirable to support the T2DM self-management process. Second, a set of eleven functional features and sub-features was developed from the research on the SDOH challenges of low income African American persons with T2DM. The T2DM apps were then compared in a two-stage process using the two sets of criteria. Because many of the criteria in the second set involved social support, only those apps that have some form of social functionality were included in the second stage comparison. Results. The results of the two comparisons are presented as two matrices comparing each app with each criterion and sub-criterion. None of the apps in stage one contained all the general functions suggested in the literature, though several come close. In stage two, most apps had few or none of the focused forms of social support for self-management capabilities of interest. Conclusions. Social capabilities of existing T2DM apps seemed based on the unconstrained social network models used in general social network media (e.g., Facebook, Twitter, Instagram). However, the needs expressed from the low-income communit","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126715336","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting protein secondary structure by an ensemble through feature-based accuracy estimation","authors":"Spencer Krieger, J. Kececioglu","doi":"10.1145/3388440.3412425","DOIUrl":"https://doi.org/10.1145/3388440.3412425","url":null,"abstract":"Protein secondary structure prediction is a fundamental task in computational biology, basic to many bioinformatics workflows, with a diverse collection of tools currently available. An approach from machine learning with the potential to capitalize on such a collection is ensemble prediction, which runs multiple predictors and combines their predictions into one, output by the ensemble. We conduct a thorough study of seven different approaches to ensemble secondary structure prediction, several of which are novel, and show we can indeed obtain an ensemble method that significantly exceeds the accuracy of individual state-of-the-art tools. The best approaches build on a recent technique known as feature-based accuracy estimation, which estimates the unknown true accuracy of a prediction, here using features of both the prediction output and the internal state of the prediction method. In particular, a hybrid approach to ensemble prediction that leverages accuracy estimation is now the most accurate method currently available: on average over standard CASP and PDB benchmarks, it exceeds the state-of-the-art Q3 accuracy for 3-state prediction by nearly 4%, and exceeds the Q8 accuracy for 8-state prediction by more than 8%. A preliminary implementation of our approach to ensemble protein secondary structure prediction, in a new tool we call Ssylla, is available free for non-commercial use at ssylla.cs.arizona.edu.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"288 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133640243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Generalized Robinson-Foulds Distance for Clonal Trees, Mutation Trees, and Phylogenetic Trees and Networks","authors":"M. Llabrés, F. Rosselló, G. Valiente","doi":"10.1145/3388440.3412479","DOIUrl":"https://doi.org/10.1145/3388440.3412479","url":null,"abstract":"Cancer evolution is often modeled by clonal trees (whose nodes are labeled by multiple somatic mutations) or mutation trees (where nodes are labeled by single somatic mutations). Clonal trees are generated from sequence data with different computational methods that may produce different clone phylogenies, rendering their analysis and comparison necessary to infer mutation order and clone origin during tumor progression. In this paper, we present a distance metric for multi-labeled trees that generalizes the Robinson-Foulds distance for phylogenetic trees, allows for a similarity assessment at much higher resolution, and can be applied to trees and networks with different sets of node labels. The generalized Robinson-Foulds distance can be computed in time quadratic in the size of the input multisets of multisets of node labels, and is a metric for clonal trees, mutation trees, phylogenetic trees, and several classes of phylogenetic networks.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129720259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Linearization of Ancestral Genomes with Duplicated Genes","authors":"P. Avdeyev, M. Alekseyev","doi":"10.1145/3388440.3412484","DOIUrl":"https://doi.org/10.1145/3388440.3412484","url":null,"abstract":"One of the key computational problems in comparative genomics is the reconstruction of genomes of ancestral species based on genomes of extant species. Since most dramatic changes in genomic architectures are caused by genome rearrangements, this problem is often posed as minimization of the number of genome rearrangements between extant and ancestral genomes. The base case of three given genomes is known as the genome median problem. Whole-genome duplications (WGDs) represent yet another type of dramatic evolutionary events and inspire the reconstruction of preduplicated ancestral genomes, referred to as the genome halving problem. Reconstruction of the preduplicated or median genomes consisting of linear chromosomes for given linear genomes is known to be intractable. There exist efficient methods for solving a relaxed version of this problem, where the preduplicated or median genomes are allowed to have circular chromosomes. Previously we proposed a method for construction of an approximate solution to the genome median problem from a solution to the relaxed version, and proved a bound on its approximation error. In this study, we extend the proposed method for constructing an approximate solution to the genome halving problem. We also extend the originally proposed method to the genome median problem for genomes with duplicated genes.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"70 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133727571","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Population-scale Genomic Data Augmentation Based on Conditional Generative Adversarial Networks","authors":"Junjie Chen, M. Mowlaei, Xinghua Shi","doi":"10.1145/3388440.3412475","DOIUrl":"https://doi.org/10.1145/3388440.3412475","url":null,"abstract":"Although next generation sequencing technologies have made it possible to quickly generate a large collection of sequences, current genomic data still suffer from small data sizes, imbalances, and biases due to various factors including disease rareness, test affordability, and concerns about privacy and security. In order to address these limitations of genomic data, we develop a Population-scale Genomic Data Augmentation based on Conditional Generative Adversarial Networks (PG-cGAN) to enhance the amount and diversity of genomic data by transforming samples already in the data rather than collecting new samples. Both the generator and discriminator in the PG-CGAN are stacked with convolutional layers to capture the underlying population structure. Our results for augmenting genotypes in human leukocyte antigen (HLA) regions showed that PC-cGAN can generate new genotypes with similar population structure, variant frequency distributions and LD patterns. Since the input for PC-cGAN is the original genomic data without assumptions about prior knowledge, it can be extended to enrich many other types of biomedical data and beyond.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"59 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124672307","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-omics data integration in the Cloud: Analysis of Statistically Significant Associations Between Clinical and Molecular Features in Breast Cancer","authors":"Kawther Abdilleh, Boris Aguilar, J. Thomson","doi":"10.1145/3388440.3414917","DOIUrl":"https://doi.org/10.1145/3388440.3414917","url":null,"abstract":"Breast Cancers are among the most common forms of cancers impacting women with over 1 million diagnoses every year worldwide. They are complex cancers characterized by distinct clinical outcomes, morphological and molecular features. As high-throughput technologies generating data at the mRNA and protein levels become cheaper and more accessible, researchers are now able to study these entities in concert with clinical features to gain a more holistic picture of Breast Cancer and other complex diseases. In this poster, we aimed at identifying the concordance or discordance of mRNA and protein expressions that are significantly associated with Breast Cancer histological subtypes and other relevant clinical features. We employed a novel cloud-based approach to analyze these statistical associations using available genomic, proteomic, and clinical cancer data on the Google Cloud through the ISB-CGC, one of the National Cancer Institute's (NCI) Cloud Resources. Our results indicate that, considering all available clinical features, a considerable number of molecules (genes and proteins) are significantly associated with the Breast Cancer histological subtypes of infiltrating ductal carcinoma and infiltrating lobular carcinoma, two common forms associated with invasive Breast Cancer. Moreover, statistically significant associations were overrepresented for molecules involved in PI3K/AKT signaling, negative regulation of the PI3K/AKT network and extra-nuclear estrogen signaling. Taken together, these results demonstrate how powerful cloud-based analytics can be in identifying novel molecular relationships relevant for Breast Cancer. text here.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"31 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116647856","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Link Analysis to Discover Insights from Structured and Unstructured Data on COVID-19","authors":"Ying Zhao, Charles C. Zhou","doi":"10.1145/3388440.3415990","DOIUrl":"https://doi.org/10.1145/3388440.3415990","url":null,"abstract":"SARS-CoV-2, the deadly and novel virus, which has caused a worldwide pandemic and drastic loss of human lives and economic activities. An open data set called the COVID-19 Open Research Dataset or CORD-19 contains large set full text scientific literature on SARS-CoV-2. The Next Strain consists of a database of SARS-CoV-2 viral genomes from 12/3/2019. We applied unique information mining method named lexical link analysis (LLA) to answer the call to action and help the science community answer high-priority scientific questions related to SARS-CoV-2. We first text-mined the CORD-19. We also data-mined the next strain database. Finally, we linked two databases: The linked databases and information can be used to discover the insights and help the research community to address high-priority questions related to the SARS-CoV-2's genetics, tests, and prevention. For example, we showed the clusters of COVID-19 cases that are consistent in terms of clinical symptoms (unstructured text descriptions from CORD-19) and genomics data (structured data from Next Strain). The genomics difference of clades A1 and A2 as shown in the Next Strain mapping may be the causes for clinical symptoms difference that are also grouped into two using the LLA method: Cases in the west coast in the United States are similar to the ones in Asia, while the more contagious and virulent ones in the east Coast in the United States are similar to ones in Europe.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128270112","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep Ranking in Template-free Protein Structure Prediction","authors":"Xiao Chen, N. Akhter, Zhiye Guo, Tianqi Wu, Jie Hou, Amarda Shehu, Jianlin Cheng","doi":"10.1145/3388440.3412469","DOIUrl":"https://doi.org/10.1145/3388440.3412469","url":null,"abstract":"The road to the discovery of the biological activities of a protein molecule in the cell goes through knowledge of its three-dimensional, biologically-active structure(s). Current evidence suggests significant regions of the protein universe are inaccessible by either wet-laboratory structure determination or homology modeling. While great progress has been made by computational approaches in elucidating dark regions of the proteome, inherent challenges remain. In this paper, we advance research on addressing one such a challenge known as model (quality) assessment. In essence, the task involves discriminating relevant structure(s) among many computed for a protein of interest. We propose a method based on deep learning and evaluate it on tertiary structures computed by a popular de-novo platform on benchmark datasets. The method uses novel protein residue-residue distance features, improved residue-residue contacts, together with other features, such as energies and model topology similarity, to estimate the quality of protein models. A detailed evaluation shows that the proposed method outperforms related ones and advances the state of the art in model assessment.","PeriodicalId":411338,"journal":{"name":"Proceedings of the 11th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123945761","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}