PaperPlayer biorxiv bioinformatics

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.278259v1?rss=1

Authors: Oakley, J. D., Sodhi, S. K., Russakoff, D. B., Choudhry, N.

Abstract:

Purpose: To evaluate the performance of a deep learning-based, fully automated, multi-class, macular fluid segmentation algorithm relative to expert annotations in a heterogeneous population of confirmed wet age-related macular degeneration (wAMD) subjects. Methods: Twenty-two swept-source optical coherence tomography (SS-OCT) volumes of the macula from 22 from different individuals with wAMD were manually annotated by two expert graders. These results were compared using cross-validation (CV) to automated segmentations using a deep learning-based algorithm encoding spatial information about retinal tissue as an additional input to the network. The algorithm detects and delineates fluid regions in the OCT data, differentiating between intra- and sub-retinal fluid (IRF, SRF), as well as fluid resulting from in serous pigment epithelial detachments (PED). Standard metrics for fluid detection and quantification were used to evaluate performance. Results: The per slice receiver operating characteristic (ROC) area under the curves (AUCs) for each of these fluid types were 0.90, 0.94 and 0.94 for IRF, SRF and PED, respectively. Per volume results were 0.94 and 0.88 for IRF and PED (SRF being present in all cases). The correlation of fluid volume between the expert graders and the algorithm were 0.99 for IRF, 0.99 for SRF and 0.82 for PED. Conclusions: Automated, deep learning-based segmentation is able to accurately detect and quantify different macular fluid types in SS-OCT data on par with expert graders.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.278002v1?rss=1

Authors: Tabassum, N.

Abstract:

Proteins are the building-blocks of life. However with the deluge of data on protein sequences and their association with diseases, it is imperative to use computational tools to aid experimental analysis. Determination of the impact of disease-causing SNPs on protein's structure is crucial in discovering how a disease affects the functions on a fundamental level and thereafter, determining potential drug targets. SNPs associated with the genetic disease permanent neonatal diabetus mellitus (PNDM) were studied to determine their impact on human insulin. Out of 16 missense variants, eight were predicted to be deleterious. 6 of the SNPs resulted in high structural differences (RMSD > 0.9, H bonds >35). Stability changes were also determined.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.08.31.274175v1?rss=1

Authors: Gu, Z., el Bouhaddani, S., Pei, J., Houwing-Duistermaat, J., Uh, H.-W.

Abstract:

Background: Nowadays, multiple omics data are measured on the same samples in the belief that these different omics datasets represent various aspects of the underlying biological systems. Integrating these omics datasets will facilitate the understanding of the systems. For this purpose, various methods have been proposed, such as Partial Least Squares (PLS), decomposing two datasets into joint and residual subspaces. Omics data are heterogeneous, and the joint subspaces estimated in PLS contain orthogonal variations unrelated to one another. Alternatively, Two-way Orthogonal Partial Least Squares (O2PLS) captures the heterogeneity by introducing the orthogonal subspaces and better estimates the joint subspaces. However, the latent components spanning the joint subspaces in O2PLS are linear combinations of all variables, while the interests of domain experts might be in a small subset. To obtain sparsity, we extend O2PLS to Group Sparse O2PLS (GO2PLS) that performs feature selection. Furthermore, features in the data often have group structures, and incorporating this information might improve the reliability of the selection procedure. Results: The simulation study showed that introducing sparsity improved the performance concerning feature selection. Furthermore, incorporating group structures increased the precision and power of the feature selection procedure. GO2PLS performed optimally in terms of accuracy of joint score estimation, joint loading estimation, and feature selection. For illustration, we applied GO2PLS to datasets from two studies: CVON-DOSIS (small case-control study) and TwinsUK (population study). In the first, by integrating regulomics and transcriptomics data, joint components using GO2PLS discriminated cardiomyopathy patients and controls better than PCA and PLS. Genes selected based on the regulatory regions and transcripts appeared to be relevant to heart muscle disease. In the second, we incorporated the information on the group structures of the methylation CpG sites when integrating the methylation dataset with the glycomics data. The selected methylation groups turned out to be relevant to the immune system, in which glycans play an important role. Conclusions: GO2PLS is an efficient approach for integrating two heterogeneous omics datasets to gain insights. It performs feature selection in both datasets, enhanced by incorporating known group structures, thereby resulting in a small subset of features for further investigation.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.08.31.273458v1?rss=1

Authors: Li, B., Veturi, Y., Verma, A., Bradford, Y., Daar, E. S., Gulick, R. M., Riddler, S. A., Robbins, G. K., Lennox, J. L., Haas, D. W., Ritchie, M. D.

Abstract:

As a type of relatively new methodology, the transcriptome-wide association study (TWAS) has gained interest due to capacity for gene-level association testing. However, the development of TWAS has outpaced statistical evaluation of TWAS gene prioritization performance. Current TWAS methods vary in underlying biological assumptions about tissue specificity of transcriptional regulatory mechanisms. In a previous study from our group, this may have affected whether TWAS methods better identified associations in single tissues versus multiple tissues. We therefore designed simulation analyses to examine how the interplay between particular TWAS methods and tissue specificity of gene expression affects power and type I error rates for gene prioritization. We found that cross-tissue identification of expression quantitative trait loci (eQTLs) improved TWAS power. Single-tissue TWAS (i.e., PrediXcan) had robust power to identify genes expressed in single tissues, but, had high false positive rates for genes that are expressed in multiple tissues. Cross-tissue TWAS (i.e., UTMOST) had overall equal or greater power and controlled type I error rates for genes expressed in multiple tissues. Based on these simulation results, we applied a tissue specificity-aware TWAS (TSA-TWAS) analytic framework to look for gene-based associations with pre-treatment laboratory values from AIDS Clinical Trial Group (ACTG) studies. We replicated several proof-of-concept transcriptionally regulated gene-trait associations, including UGT1A1 (encoding bilirubin uridine diphosphate glucuronosyl transferase enzyme) and total bilirubin levels (p = 3.59e-12), and CETP (cholesteryl ester transfer protein) with high-density lipoprotein cholesterol (p = 4.49e-12). We also identified several novel genes associated with metabolic and virologic traits, as well as pleiotropic genes that linked plasma viral load, absolute basophil count, and/or triglyceride levels. By highlighting the advantages of different TWAS methods, our simulation study promotes a tissue specificity-aware TWAS analytic framework that revealed novel aspects of HIV-related traits.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.08.30.274217v1?rss=1

Authors: Yang, Q., Warnow, T.

Abstract:

PASTA is a method for estimating alignments and trees that has been able to provide excellent accuracy on large sequence datasets. By design, PASTA operates using iteration, in which the tree from the previous iteration is used to inform a divide-and-conquer strategy during which a new alignment is computed on the sequence dataset, and then a new maximum likelihood tree is estimated on the new alignment. In its default setting, PASTA runs for three iterations and returns that alignment/tree pair from the last iteration. Here we use both biological and simulated nucleotide datasets to show that returning the alignment/tree pair that has the best maximum likelihood score improves on the default usage.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.277335v1?rss=1

Authors: Do, C. P. N., Nguyen, T. H., Vo, U. T. M., Le, L.

Abstract:

Influenza virus A is a significant agent involved in the outbreak of worldwide epidemics, causing millions of fatalities around the world by respiratory diseases and seasonal illness. Many projects had been conducting to investigate recovered infected patients for therapeutic vaccines that have broad-spectrum activity. With the aid of the computational approach in biology, the designation for a vaccine model is more accessible. We developed an in silico protocol called i BRAB to design a broad-reactive Fab on a wide range of influenza A virus. The Fab model was constructed based on sequences and structure of available broad-spectrum Abs or Fabs against a wide range of H1N1 influenza A virus. As a result, the proposed Fab model followed iBRAB has good binding affinity over 27 selected HA of different strains of H1 influenza A virus, including wild-type and mutated ones. The examination also took by computational tools to fasten the procedure. This protocol could be applied for a fast designed therapeutic vaccine against different types of threats.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.277251v1?rss=1

Authors: Chen, B.-N., Song, P., Chen, M.-C., Hong, M.-C.

Abstract:

Endosymbiosis between dinoflagellates and cnidarian hosts first occurred more than 200 million years ago; however, symbiosis-specific genes and cellular processes involved in the establishment, maintenance, and breakdown of endosymbiosis remain unclear. Therefore, this study aimed to identify the zooxanthellal genes associated with the aforementioned biological processes during endosymbiosis in Aiptasia-Symbiodinium endosymbionts. Here, zooxanthellae isolates were treated with 0.02% SDS to decrease potential host RNA contamination and to enhance the identification of novel symbiosis/nonsymbiosis-associated differentially expressed zooxanthellal genes through suppressive subtractive hybridization (SSH) and next-generation sequencing (NGS) methods. Consequently, among 214 symbiosis-specific transcripts identified herein that displayed identity to only 5.6% of host-derived transcripts, 64% were well-known functional genes. In the nonsymbiotic stage, 181 differentially expressed transcripts were identified, of which 64.1% belonged to well-known functional genes. BLAST revealed that 8 categories of cellular processes were significantly induced in symbiotic or nonsymbiotic zooxanthellae. Together with the results of quantitative analysis, the results revealed that photosynthesis, flagellate biosynthesis and motility, stress-induced responses, cell wall biosynthesis, starch synthesis and transport, lipid biosynthesis and metabolism, host/symbiont immune response, intercellular communication, cell growth, and cell cycle regulation were the major cellular processes occurring in symbiotic/nonsymbiotic stages. The present results provide insights into the mechanisms involved in regulating the different physiological processes in symbiotic/nonsymbiotic zooxanthellae and may guide future studies.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.277277v1?rss=1

Authors: Giles, O., Karlsson, A., Masiala, S., White, S., Cesareni, G., Perfetto, L., Mullen, J., Hughes, M., Harland, L., Malone, J.

Abstract:

Text mining is widely used within the life sciences as an evidence stream for inferring relationships between biological entities. In most cases, conventional string matching is used to identify cooccurrences of given entities within sentences. This limits the utility of text mining results, as they tend to contain significant noise due to weak inclusion criteria. We show that, in the indicative case of protein-protein interactions (PPIs), the majority of sentences containing cooccurrences (~75%) do not describe any causal relationship. We further demonstrate the feasibility of fine tuning a strong domain-specific language model, BioBERT, to analyse sentences containing cooccurrences and accurately (F1 score: 88.95%) identify functional links between proteins. These strong results come in spite of the deep complexity of the language involved, which limits the accuracy even of expert curators. We establish guidelines for best practices in data creation to this end, including an examination of inter-annotator agreement, of semisupervision, and of rules based alternatives to manual curation, and explore the potential for downstream use of the model to accelerate curation of interactions in the SIGNOR database of causal protein interactions and the IntAct database of experimental evidence for physical protein interactions.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.278465v1?rss=1

Authors: Arslan, A., GUAN, Y., Chen, X., Zhu, W., Wu, M., Zheng, M., Dill, D. L., Peltz, G., Donaldson, R., Ford, M.

Abstract:

Background: Genetic factors affecting multiple biomedical traits in mice have been identified when GWAS data, which measured responses in panels of inbred mouse strains, was analyzed using haplotype-based computational genetic mapping (HBCGM). Although this method was previously used to analyze one dataset at a time; but now, a vast amount of mouse phenotypic data is now publicly available, which could enable many more genetic discoveries. Results: HBCGM and a whole genome SNP map covering 43 inbred strains was used to analyze 8300 publicly available datasets of biomedical responses (1.52M individual datapoints) measured in panels of inbred mouse strains. As proof of concept, causative genetic factors affecting susceptibility for eye, metabolic and infectious diseases were identified when structured automated methods were used to analyze the output. One analysis identified a novel genetic effector mechanism; allelic differences within the mitochondrial targeting sequence affected the subcellular localization of a protein. We also found allelic differences within the mitochondrial targeting sequences of many murine and human proteins, and these could affect a wide range of biomedical phenotypes. Implications: These initial results indicate that genetic factors affecting biomedical responses could be identified through analysis of very large datasets, and they provide an early indication of how this type of 'augmented intelligence' can facilitate genetic discovery. Abbreviations: cSNP, codon-changing SNP; GWAS, genome-wide association study; HBCGM, haplotype-based computational genetic mapping; MTS, Mitochondrial targeting sequence.

Copy rights belong to original authors. Visit the link for more info

Link to bioRxiv paper:

http://biorxiv.org/cgi/content/short/2020.09.01.277632v1?rss=1

Authors: Bien, J., Yan, X., Simpson, L., Müller, C. L.

Abstract:

Modern high-throughput sequencing technologies provide low-cost microbiome survey data across all habitats of life at unprecedented scale. At the most granular level, the primary data consist of sparse counts of amplicon sequence variants or operational taxonomic units that are associated with taxonomic and phylogenetic group information. In this contribution, we leverage the hierarchical structure of amplicon data and propose a data-driven, parameter-free, and scalable tree-guided aggregation framework to associate microbial subcompositions with response variables of interest. The excess number of zero or low count measurements at the read level forces traditional microbiome data analysis workflows to remove rare sequencing variants or group them by a fixed taxonomic rank, such as genus or phylum, or by phylogenetic similarity. By contrast, our framework, which we call trac (tree-aggregation of compositional data), learns data-adaptive taxon aggregation levels for predictive modeling making user-defined aggregation obsolete while simultaneously integrating seamlessly into the compositional data analysis framework. We illustrate the versatility of our framework in the context of large-scale regression problems in human-gut, soil, and marine microbial ecosystems. We posit that the inferred aggregation levels provide highly interpretable taxon groupings that can help microbial ecologists gain insights into the structure and functioning of the underlying ecosystem of interest.

Copy rights belong to original authors. Visit the link for more info