Knowledge Discovery and interactive Data Mining in Bioinformatics--State-of-the-Art, future challenges and research directions

Studies of the role of a smart web for precision medicine supported by biobanking

Both the extraction of medical knowledge from data mining many patient records and from authoritative natural language text on the Internet are important for clinical decision support and biomedical research. The samples in biobanks represent a further kind of information repository of recognized increasing importance, so mechanisms being developed for a smart web for medicine should take them into account. While this paper is primarily a review of Quantum Universal Exchange Language as an XML extension to enable a future smart web for healthcare and biomedicine, it is the first time that we have discussed the connection with biobanks and the design of Quantum Universal Exchange Language's XML-like tags to support their use.

Detection of Cardiovascular Disease Risk's Level for Adults Using Naive Bayes Classifier

OBJECTIVES

The number of deaths caused by cardiovascular disease and stroke is predicted to reach 23.3 million in 2030. As a contribution to support prevention of this phenomenon, this paper proposes a mining model using a naïve Bayes classifier that could detect cardiovascular disease and identify its risk level for adults.

METHODS

The process of designing the method began by identifying the knowledge related to the cardiovascular disease profile and the level of cardiovascular disease risk factors for adults based on the medical record, and designing a mining technique model using a naïve Bayes classifier. Evaluation of this research employed two methods: accuracy, sensitivity, and specificity calculation as well as an evaluation session with cardiologists and internists. The characteristics of cardiovascular disease are identified by its primary risk factors. Those factors are diabetes mellitus, the level of lipids in the blood, coronary artery function, and kidney function. Class labels were assigned according to the values of these factors: risk level 1, risk level 2 and risk level 3.

RESULTS

The evaluation of the classifier performance (accuracy, sensitivity, and specificity) in this research showed that the proposed model predicted the class label of tuples correctly (above 80%). More than eighty percent of respondents (including cardiologists and internists) who participated in the evaluation session agree till strongly agreed that this research followed medical procedures and that the result can support medical analysis related to cardiovascular disease.

CONCLUSIONS

The research showed that the proposed model achieves good performance for risk level detection of cardiovascular disease.

Visual analytics for concept exploration in subspaces of patient groups : Making sense of complex datasets with the Doctor-in-the-loop

Medical doctors and researchers in bio-medicine are increasingly confronted with complex patient data, posing new and difficult analysis challenges. These data are often comprising high-dimensional descriptions of patient conditions and measurements on the success of certain therapies. An important analysis question in such data is to compare and correlate patient conditions and therapy results along with combinations of dimensions. As the number of dimensions is often very large, one needs to map them to a smaller number of relevant dimensions to be more amenable for expert analysis. This is because irrelevant, redundant, and conflicting dimensions can negatively affect effectiveness and efficiency of the analytic process (the so-called curse of dimensionality). However, the possible mappings from high- to low-dimensional spaces are ambiguous. For example, the similarity between patients may change by considering different combinations of relevant dimensions (subspaces). We demonstrate the potential of subspace analysis for the interpretation of high-dimensional medical data. Specifically, we present SubVIS, an interactive tool to visually explore subspace clusters from different perspectives, introduce a novel analysis workflow, and discuss future directions for high-dimensional (medical) data analysis and its visual exploration. We apply the presented workflow to a real-world dataset from the medical domain and show its usefulness with a domain expert evaluation.

Interactive machine learning for health informatics: when do we need the human-in-the-loop?

Machine learning (ML) is the fastest growing field in computer science, and health informatics is among the greatest challenges. The goal of ML is to develop algorithms which can learn and improve over time and can be used for predictions. Most ML researchers concentrate on automatic machine learning (aML), where great advances have been made, for example, in speech recognition, recommender systems, or autonomous vehicles. Automatic approaches greatly benefit from big data with many training sets. However, in the health domain, sometimes we are confronted with a small number of data sets or rare events, where aML-approaches suffer of insufficient training samples. Here interactive machine learning (iML) may be of help, having its roots in reinforcement learning, preference learning, and active learning. The term iML is not yet well used, so we define it as “algorithms that can interact with agents and can optimize their learning behavior through these interactions, where the agents can also be human.” This “human-in-the-loop” can be beneficial in solving computationally hard problems, e.g., subspace clustering, protein folding, or k-anonymization of health data, where human expertise can help to reduce an exponential search space through heuristic selection of samples. Therefore, what would otherwise be an NP-hard problem, reduces greatly in complexity through the input and the assistance of a human agent involved in the learning phase.

Interactive knowledge discovery with the doctor-in-the-loop: a practical example of cerebral aneurysms research

Established process models for knowledge discovery find the domain-expert in a customer-like and supervising role. In the field of biomedical research, it is necessary to move the domain-experts into the center of this process with far-reaching consequences for both their research output and the process itself. In this paper, we revise the established process models for knowledge discovery and propose a new process model for domain-expert-driven interactive knowledge discovery. Furthermore, we present a research infrastructure which is adapted to this new process model and demonstrate how the domain-expert can be deeply integrated even into the highly complex data-mining process and data-exploration tasks. We evaluated this approach in the medical domain for the case of cerebral aneurysms research.

Portal of medical data models: information infrastructure for medical research and healthcare

INTRODUCTION

Information systems are a key success factor for medical research and healthcare. Currently, most of these systems apply heterogeneous and proprietary data models, which impede data exchange and integrated data analysis for scientific purposes. Due to the complexity of medical terminology, the overall number of medical data models is very high. At present, the vast majority of these models are not available to the scientific community. The objective of the Portal of Medical Data Models (MDM, https://medical-data-models.org) is to foster sharing of medical data models.

METHODS

MDM is a registered European information infrastructure. It provides a multilingual platform for exchange and discussion of data models in medicine, both for medical research and healthcare. The system is developed in collaboration with the University Library of Münster to ensure sustainability. A web front-end enables users to search, view, download and discuss data models. Eleven different export formats are available (ODM, PDF, CDA, CSV, MACRO-XML, REDCap, SQL, SPSS, ADL, R, XLSX). MDM contents were analysed with descriptive statistics.

RESULTS

MDM contains 4387 current versions of data models (in total 10,963 versions). 2475 of these models belong to oncology trials. The most common keyword (n = 3826) is 'Clinical Trial'; most frequent diseases are breast cancer, leukemia, lung and colorectal neoplasms. Most common languages of data elements are English (n = 328,557) and German (n = 68,738). Semantic annotations (UMLS codes) are available for 108,412 data items, 2453 item groups and 35,361 code list items. Overall 335,087 UMLS codes are assigned with 21,847 unique codes. Few UMLS codes are used several thousand times, but there is a long tail of rarely used codes in the frequency distribution.

DISCUSSION

Expected benefits of the MDM portal are improved and accelerated design of medical data models by sharing best practice, more standardised data models with semantic annotation and better information exchange between information systems, in particular Electronic Data Capture (EDC) and Electronic Health Records (EHR) systems. Contents of the MDM portal need to be further expanded to reach broad coverage of all relevant medical domains. Database URL: https://medical-data-models.org.

Discrimination Power of Polynomial-Based Descriptors for Graphs by Using Functional Matrices

In this paper, we study the discrimination power of graph measures that are based on graph-theoretical matrices. The paper generalizes the work of [M. Dehmer, M. Moosbrugger. Y. Shi, Encoding structural information uniquely with polynomial-based descriptors by employing the Randić matrix, Applied Mathematics and Computation, 268(2015), 164–168]. We demonstrate that by using the new functional matrix approach, exhaustively generated graphs can be discriminated more uniquely than shown in the mentioned previous work.

General overview on structure prediction of twilight-zone proteins

Protein structure prediction from amino acid sequence has been one of the most challenging aspects in computational structural biology despite significant progress in recent years showed by critical assessment of protein structure prediction (CASP) experiments. When experimentally determined structures are unavailable, the predictive structures may serve as starting points to study a protein. If the target protein consists of homologous region, high-resolution (typically <1.5 Å) model can be built via comparative modelling. However, when confronted with low sequence similarity of the target protein (also known as twilight-zone protein, sequence identity with available templates is less than 30%), the protein structure prediction has to be initiated from scratch. Traditionally, twilight-zone proteins can be predicted via threading or ab initio method. Based on the current trend, combination of different methods brings an improved success in the prediction of twilight-zone proteins. In this mini review, the methods, progresses and challenges for the prediction of twilight-zone proteins were discussed.

Integrated web visualizations for protein-protein interaction databases

Background

Understanding living systems is crucial for curing diseases. To achieve this task we have to understand biological networks based on protein-protein interactions. Bioinformatics has come up with a great amount of databases and tools that support analysts in exploring protein-protein interactions on an integrated level for knowledge discovery. They provide predictions and correlations, indicate possibilities for future experimental research and fill the gaps to complete the picture of biochemical processes. There are numerous and huge databases of protein-protein interactions used to gain insights into answering some of the many questions of systems biology. Many computational resources integrate interaction data with additional information on molecular background. However, the vast number of diverse Bioinformatics resources poses an obstacle to the goal of understanding. We present a survey of databases that enable the visual analysis of protein networks.

Results

We selected M =10 out of N =53 resources supporting visualization, and we tested against the following set of criteria: interoperability, data integration, quantity of possible interactions, data visualization quality and data coverage. The study reveals differences in usability, visualization features and quality as well as the quantity of interactions. StringDB is the recommended first choice. CPDB presents a comprehensive dataset and IntAct lets the user change the network layout. A comprehensive comparison table is available via web. The supplementary table can be accessed on http://tinyurl.com/PPI-DB-Comparison-2015.

Conclusions

Only some web resources featuring graph visualization can be successfully applied to interactive visual analysis of protein-protein interaction. Study results underline the necessity for further enhancements of visualization integration in biochemical analysis tools. Identified challenges are data comprehensiveness, confidence, interactive feature and visualization maturing.