Consistent probabilistic outputs for protein function prediction

Background:

The wide availability of genome-scale data for several organisms has stimulated interest in computational approaches to gene function prediction. Diverse machine learning methods have been applied to unicellular organisms with some success, but few have been extensively tested on higher level, multicellular organisms. A recent mouse function prediction project (MouseFunc) brought together nine bioinformatics teams applying a diverse array of methodologies to mount the first large-scale effort to predict gene function in the laboratory mouse.

Results:

In this paper, we describe our contribution to this project, an ensemble framework based on the support vector machine that integrates diverse datasets in the context of the Gene Ontology hierarchy. We carry out a detailed analysis of the performance of our ensemble and provide insights into which methods work best under a variety of prediction scenarios. In addition, we applied our method to Saccharomyces cerevisiae and have experimentally confirmed functions for a novel mitochondrial protein.

Conclusion:

Our method consistently performs among the top methods in the MouseFunc evaluation. Furthermore, it exhibits good classification performance across a variety of cellular processes and functions in both a multicellular organism and a unicellular organism, indicating its ability to discover novel biology in diverse settings.

     

Update in research and methods in proteomics and bioinformatics

The 3rd International Conference on Proteomics & Bioinformatics (Proteomics 2013)

Philadelphia, PA, USA, 15–17 July 2013

The Third International Conference on Proteomics & Bioinformatics (Proteomics 2013) was sponsored by the OMICS group and was organized in order to strengthen the future of proteomics science by bringing together professionals, researchers and scholars from leading universities across the globe. The main topics of this conference included the integration of novel platforms in data analysis, the use of a systems biology approach, different novel mass spectrometry platforms and biomarker discovery methods. The conference was divided into proteomic methods and research interests. Among these two categories, interactions between methods in proteomics and bioinformatics, as well as other research methodologies, were discussed. Exceptional topics from the keynote forum, oral presentations and the poster session have been highlighted. The topics range from new techniques for analyzing proteomics data, to new models designed to help better understand genetic variations to the differences in the salivary proteomes of HIV-infected patients.     

Applications and Challenges of Machine Learning Methods in Alzheimer's Disease Multi-Source Data Analysis

BackgroundRecent development in neuroimaging and genetic testing technologies have made it possible to measure pathological features associated with Alzheimer''s disease (AD) in vivo. Mining potential molecular markers of AD from high-dimensional, multi-modal neuroimaging and omics data will provide a new basis for early diagnosis and intervention in AD. In order to discover the real pathogenic mutation and even understand the pathogenic mechanism of AD, lots of machine learning methods have been designed and successfully applied to the analysis and processing of large-scale AD biomedical data.ObjectiveTo introduce and summarize the applications and challenges of machine learning methods in Alzheimer''s disease multi-source data analysis.MethodsThe literature selected in the review is obtained from Google Scholar, PubMed, and Web of Science. The keywords of literature retrieval include Alzheimer''s disease, bioinformatics, image genetics, genome-wide association research, molecular interaction network, multi-omics data integration, and so on.ConclusionThis study comprehensively introduces machine learning-based processing techniques for AD neuroimaging data and then shows the progress of computational analysis methods in omics data, such as the genome, proteome, and so on. Subsequently, machine learning methods for AD imaging analysis are also summarized. Finally, we elaborate on the current emerging technology of multi-modal neuroimaging, multi-omics data joint analysis, and present some outstanding issues and future research directions.     

Global profiling of protein complexes: current approaches and their perspective in biomedical research

Introduction: Despite the rapid evolution of proteomic methods, protein interactions and their participation in protein complexes – an important aspect of their function – has rarely been investigated on the proteome-wide level. Disease states, such as muscular dystrophy or viral infection, are induced by interference in protein-protein interactions within complexes. The purpose of this review is to describe the current methods for global complexome analysis and to critically discuss the challenges and opportunities for the application of these methods in biomedical research.

Areas covered: We discuss advancements in experimental techniques and computational tools that facilitate profiling of the complexome. The main focus is on the separation of native protein complexes via size exclusion chromatography and gel electrophoresis, which has recently been combined with quantitative mass spectrometry, for a global protein-complex profiling. The development of this approach has been supported by advanced bioinformatics strategies and fast and sensitive mass spectrometers that have allowed the analysis of whole cell lysates. The application of this technique to biomedical research is assessed, and future directions are anticipated.

Expert commentary: The methodology is quite new, and has already shown great potential when combined with complementary methods for detection of protein complexes.     

MACBenAbim: A Multi-platform Mobile Application for searching keyterms in Computational Biology and Bioinformatics

Computational biology and bioinformatics are gradually gaining grounds in Africa and other developing nations of the world. However, in these countries, some of the challenges of computational biology and bioinformatics education are inadequate infrastructures, and lack of readily-available complementary and motivational tools to support learning as well as research. This has lowered the morale of many promising undergraduates, postgraduates and researchers from aspiring to undertake future study in these fields. In this paper, we developed and described MACBenAbim (Multi-platform Mobile Application for Computational Biology and Bioinformatics), a flexible user-friendly tool to search for, define and describe the meanings of keyterms in computational biology and bioinformatics, thus expanding the frontiers of knowledge of the users. This tool also has the capability of achieving visualization of results on a mobile multi-platform context.

Availability

MACBenAbim is available from the authors for non-commercial purposes.     

In silico method for systematic analysis of feature importance in microRNA-mRNA interactions

Background  

MicroRNA (miRNA), which is short non-coding RNA, plays a pivotal role in the regulation of many biological processes and affects the stability and/or translation of mRNA. Recently, machine learning algorithms were developed to predict potential miRNA targets. Most of these methods are robust but are not sensitive to redundant or irrelevant features. Despite their good performance, the relative importance of each feature is still unclear. With increasing experimental data becoming available, research interest has shifted from higher prediction performance to uncovering the mechanism of microRNA-mRNA interactions.     

Prospective sustainability assessment: the case of wood in automotive applications

Purpose

The introduction of renewable materials into automotive applications is perceived as an innovative lightweight solution. Wood-based materials are advantageous in that they have potentially lower environmental impacts as compared with other materials such as steel. However, using wood per se does not automatically ensure more sustainability. Few prospective sustainability assessment methods or studies on the use of wood-based materials in automotive applications have been carried out, although these are needed to reduce unintended, negative sustainability effects and to support sustainable oriented research and innovation. Therefore, this study was conducted to assess the potential sustainability effects and consequences of introducing a wood-based component into an automotive application.

Methods

A combination of methods was used to analyze the potential sustainability effects when introducing wood into automotive applications. This prospective life cycle sustainability analysis solely relied on secondary data. The environmental impacts were analyzed using a simplified environmental life cycle assessment on the product level. A multi-regional input-output-based assessment was conducted to model the country-specific environmental and socioeconomic consequences. The potential shift in social risks and opportunities on a national scale was analyzed by conducting a generic social life cycle assessment. Various aspects of each approach differ, with each providing a specific perspective of the system under study.

Results and discussion

The results indicate that implementing wood into automotive application can have environmental, social, and economic benefits, according to most of the indicators analyzed. Mostly due to the product weight reduction due to the use of a wood-based component, the results show that environmental impacts decrease. Some possible consequences of using wood-based materials are increased value added and increasing the number of jobs in European countries. Similarly, the social risks and opportunities are shifted from countries all over the world to European countries, which perform better than developing countries according to several indicators. However, some indicators, such as migrant acceptance or local supplier quantity, perform better in the current situation.

Conclusions

The presented case study is particularly notable, because the results clearly indicate the advantages of using wood-based materials in automotive applications, although the application of such relatively holistic and complex approaches often may lead to rather indifferent pictures. Policy makers, researchers, and companies can apply this combination of methods that rely solely on generic data to obtain both feasible and informative results. These methods also allow users to link the product level assessment with a regional and social perspective and screen critical topics to support sustainability research and innovation.

     

Demystifying computer science for molecular ecologists

In this age of data‐driven science and high‐throughput biology, computational thinking is becoming an increasingly important skill for tackling both new and long‐standing biological questions. However, despite its obvious importance and conspicuous integration into many areas of biology, computer science is still viewed as an obscure field that has, thus far, permeated into only a few of the biology curricula across the nation. A national survey has shown that lack of computational literacy in environmental sciences is the norm rather than the exception [Valle & Berdanier (2012) Bulletin of the Ecological Society of America, 93, 373–389]. In this article, we seek to introduce a few important concepts in computer science with the aim of providing a context‐specific introduction aimed at research biologists. Our goal was to help biologists understand some of the most important mainstream computational concepts to better appreciate bioinformatics methods and trade‐offs that are not obvious to the uninitiated.     

Mining cancer biology through bioinformatic analysis of proteomic data

ABSTRACT

Introduction: Discovery proteomics for cancer research generates complex datasets of diagnostic, prognostic, and therapeutic significance in human cancer. With the advent of high-resolution mass spectrometers, able to identify thousands of proteins in complex biological samples, only the application of bioinformatics can lead to the interpretation of data which can be relevant for cancer research.

Areas covered: Here, we give an overview of the current bioinformatic tools used in cancer proteomics. Moreover, we describe their applications in cancer proteomics studies of cell lines, serum, and tissues, highlighting recent results and critically evaluating their outcomes.

Expert opinion: The use of bioinformatic tools is a fundamental step in order to manage the large amount of proteins (from hundreds to thousands) that can be identified and quantified in a cancer biological samples by proteomics. To handle this challenge and obtain useful data for translational medicine, it is important the combined use of different bioinformatic tools. Moreover, a particular attention to the global experimental design, and the integration of multidisciplinary skills are essential for best setting of tool parameters and best interpretation of bioinformatics output.     

Evolving research trends in bioinformatics

The cross-disciplinary nature of bioinformatics entails co-evolution with other biomedical disciplines, whereby some bioinformatics applications become popular in certain disciplines and, in turn, these disciplines influence the focus of future bioinformatics development efforts. We observe here that the growth of computational approaches within various biomedical disciplines is not merely a reflection of a general extended usage of computers and the Internet, but due to the production of useful bioinformatics databases and methods for the rest of the biomedical scientific community. We have used the abstracts stored both in the MEDLINE database of biomedical literature and in NIH-funded project grants, to quantify two effects. First, we examine the biomedical literature as a whole and find that the use of computational methods has become increasingly prevalent across biomedical disciplines over the past three decades, while use of databases and the Internet have been rapidly increasing over the past decade. Second, we study the recent trends in the use of bioinformatics topics. We observe that molecular sequence databases are a widely adopted contribution in biomedicine from the field of bioinformatics, and that microarray analysis is one of the major new topics engaged by the bioinformatics community. Via this analysis, we were able to identify areas of rapid growth in the use of informatics to aid in curriculum planning, development of computational infrastructure and strategies for workforce education and funding.     

Lost in the crowd: identifying targetable MHC class I neoepitopes for cancer immunotherapy

ABSTRACT

Introduction: The recent development of checkpoint blockade immunotherapy for cancer has led to impressive clinical results across multiple tumor types. There is mounting evidence that immune recognition of tumor derived MHC class I (MHC-I) restricted epitopes bearing cancer specific mutations and alterations is a crucial mechanism in successfully triggering immune-mediated tumor rejection. Therapeutic targeting of these cancer specific epitopes (neoepitopes) is emerging as a promising opportunity for the generation of personalized cancer vaccines and adoptive T cell therapies. However, one major obstacle limiting the broader application of neoepitope based therapies is the difficulty of selecting highly immunogenic neoepitopes among the wide array of presented non-immunogenic HLA ligands derived from self-proteins.

Areas covered: In this review, we present an overview of the MHC-I processing and presentation pathway, as well as highlight key areas that contribute to the complexity of the associated MHC-I peptidome. We cover recent technological advances that simplify and optimize the identification of targetable neoepitopes for cancer immunotherapeutic applications.

Expert commentary: Recent advances in computational modeling, bioinformatics, and mass spectrometry are unlocking the underlying mechanisms governing antigen processing and presentation of tumor-derived neoepitopes.     

Recruiting machine learning methods for molecular simulations of proteins

Abstract

Molecular dynamics (MD) simulations are critical to understanding the movements of proteins in time. Yet, MD simulations are limited due to the availability of high-resolution protein structures, accuracy of the underlying force-field, computational expense, and difficulty in analysing big data-sets. Machine learning algorithms are now routinely used to circumvent many of these limitations and computational biophysicists are continuously making progress in developing novel applications. Here, we discuss some of these methods, varying from traditional dimensionality reduction approaches to more recent abstractions such as transfer learning and reinforcement learning, and how they have been used to deal with the challenges in MD. We conclude with the prospective issues in the application of machine learning methods in MD, to increase accuracy and efficiency of protein dynamics studies in general.     

Dependency parsing of biomedical text with BERT

Background

Entity normalization is an important information extraction task which has gained renewed attention in the last decade, particularly in the biomedical and life science domains. In these domains, and more generally in all specialized domains, this task is still challenging for the latest machine learning-based approaches, which have difficulty handling highly multi-class and few-shot learning problems. To address this issue, we propose C-Norm, a new neural approach which synergistically combines standard and weak supervision, ontological knowledge integration and distributional semantics.

Results

Our approach greatly outperforms all methods evaluated on the Bacteria Biotope datasets of BioNLP Open Shared Tasks 2019, without integrating any manually-designed domain-specific rules.

Conclusions

Our results show that relatively shallow neural network methods can perform well in domains that present highly multi-class and few-shot learning problems.

     

Detecting antimicrobial peptides by exploring the mutual information of their sequences

Abstract

The rise of antibiotic resistance in pathogenic bacteria is a growing concern for every part of the world. The present study shows the prediction efficiency of mutual information for the classification of antimicrobial peptides. The proven role of antimicrobial peptides (AMPs) to fight against multidrug-resistant pathogens and AMP’s low toxic properties laid the foundation of computational methods to play their role in detecting AMPs from non-AMPs. Mutual information vectors (MIV) were created for AMP/non-AMP sequences and then fed to different machine learning classifiers out of which a random forest (RF) classifier showed best results for predicting AMPs. Random forest classifiers were evaluated on benchmark datasets by 10-fold cross-validation. The proposed MIV-RF method showed better prediction accuracy, MCC (Matthews correlation coefficient), and AUC-ROC (Area Under The Curve-Receiver Operating Characteristics) than available methods for detecting AMPs.

Communicated by Ramaswamy H. Sarma     

Is mainstream LCA linear?

Purpose

It is frequently mentioned in literature that LCA is linear, without a proof, or even without a clear definition of the criterion for linearity. Here we study the meaning of the term linear, and in relation to that, the question if LCA is indeed linear.

Methods

We explore the different meanings of the term linearity in the context of mathematical models. This leads to a distinction between linear functions, homogeneous functions, homogenous linear functions, bilinear functions, and multilinear functions. Each of them is defined in accessible terms and illustrated with examples.

Results

We analyze traditional, matrix-based, LCA, and conclude that LCA is not linear in any of the senses defined.

Discussion and conclusions

Despite the negative answer to the research question, there are many respects in which LCA can be regarded to be, at least to some extent, linear. We discuss a few of such cases. We also discuss a few practical implications for practitioners of LCA and for developers of new methods for LCI and LCIA.

     

我国护理人员绩效考核的发展研究

目的 了解我国护理人员绩效考核的现状及其发展趋势,分析目前我国护理人员绩效考核体系的结构,为建立完善、有效的护理人员考核体系提供思路。 方法 检索中文数据库中核心期刊上公开发表的关于护理人员绩效考核的论文,对其进行描述性和半定量分析。 结果 我国护理人员的绩效考核经历了从经验性介绍到应用性研究的过程,研究方法逐渐由描述性向方法学研究发展。结论 目前国内关于护理人员考核体系的种类较多,缺乏统一的标准,考核结果在全国范围内缺乏可比性。

     

COST Action CM1201 “Biomimetic Radical Chemistry”: free radical chemistry successfully meets many disciplines

Abstract

The COST Action CM1201 “Biomimetic Radical Chemistry” has been active since December 2012 for 4 years, developing research topics organized into four working groups: WG1 – Radical Enzymes, WG2 – Models of DNA damage and consequences, WG3 – Membrane stress, signalling and defenses, and WG4 – Bio-inspired synthetic strategies. International collaborations have been established among the participating 80 research groups with brilliant interdisciplinary achievements. Free radical research with a biomimetic approach has been realized in the COST Action and are summarized in this overview by the four WG leaders.     

Machine learning approaches in MALDI-MSI: clinical applications

Introduction: Despite the unquestionable advantages of Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging in visualizing the spatial distribution and the relative abundance of biomolecules directly on-tissue, the yielded data is complex and high dimensional. Therefore, analysis and interpretation of this huge amount of information is mathematically, statistically and computationally challenging.

Areas covered: This article reviews some of the challenges in data elaboration with particular emphasis on machine learning techniques employed in clinical applications, and can be useful in general as an entry point for those who want to study the computational aspects. Several characteristics of data processing are described, enlightening advantages and disadvantages. Different approaches for data elaboration focused on clinical applications are also provided. Practical tutorial based upon Orange Canvas and Weka software is included, helping familiarization with the data processing.

Expert commentary: Recently, MALDI-MSI has gained considerable attention and has been employed for research and diagnostic purposes, with successful results. Data dimensionality constitutes an important issue and statistical methods for information-preserving data reduction represent one of the most challenging aspects. The most common data reduction methods are characterized by collecting independent observations into a single table. However, the incorporation of relational information can improve the discriminatory capability of the data.