WILF 2021 - GNCS-abstract

Netpro2vec: a Graph Embedding Framework for Biomedical Applications

Mario Guarracino, Università degli Studi di Cassino e Lazio Meridionale

In this talk we will detail Netpro2vec, a method for learning features on graphs based on a neural embedding framework. It uses probability distribution representations of graphs. It uses node descriptions such as those induced by the Transition Matrix and Node Distance Distribution. Netpro2vec provides embeddings completely independent from the task and nature of the data. The framework is evaluated on synthetic and various real biomedical network datasets through a comprehensive experimental classification phase and is compared to well-known competitors.

NETME: On-the-Fly Knowledge Network Construction from Biomedical Literature

Alfredo Pulvirenti, Università degli Studi di Catania

The rapidly increasing biological literature is a key resource to automatically extract and gain knowledge concerning biological elements and their relations. Knowledge Networks are helpful tools in the context of biological knowledge discovery and modeling. In the talk I will introduce a novel system called NetME, which, starting from a set of full-texts obtained from PUBMED, through an easy-to-use web interface, interactively extracts biological elements from ontological databases and then synthesizes a network inferring relations among such elements. The results clearly show that NetME is capable of inferring comprehensive and reliable biological networks.

Classification of Sequences with Deep Artificial Neural Networks: Representation and Architectural Issues

Giosuè Lo Bosco, Università degli Studi di Palermo

DNA sequences are the basic data type that is processed to perform a generic study of biological data analysis. One key component of the biological analysis is represented by sequence classiﬁcation, a methodology that is widely used to analyze sequential data of different nature. However, its application to DNA sequences requires a proper representation of such sequences, which is still an open research problem. Machine Learning (ML) methodologies have given a fundamental contribution to the solution of the problem. Among them, recently, also Deep Neural Network (DNN) models have shown strongly encouraging results. In this study, we deal with speciﬁc classiﬁcation problems related to two biological scenarios: (A) metagenomics and (B) chromatin organization. The investigations have been carried out by considering DNA sequences as input data for the classiﬁcation methodologies. In particular, we study and test the efficacy of (1) different DNA sequence representations and (2) several Deep Learning (DL) architectures that process sequences for the solution of the related supervised classiﬁcation problems. Although developed for speciﬁc classiﬁcation tasks, we think that such architectures could be served as a suggestion for developing other DNN models that process the same kind of input.

Multi-View Data Integration of Omics Data

Angelo Ciaramella, Università degli Studi di Napoli Parthenope

We present a multi-view integration methodology (named FH-Clust) for identifying patient subgroups from different omics information (e.g., Gene Expression, Mirna Expression, Methylation). In particular, hierarchical structures of patient data are obtained in each omic (or view) and finally their topologies are merged by consensus matrix. One of the main aspects of this methodology, is the use of a measure of dissimilarity between sets of observations, by using an appropriate metric based on a fuzzy equivalence relation with Łukasiewicz valued fuzzy similarity. Several experiments and comparisons are made on real data (e.g., Glioblastoma, Prostate Cancer) to assess the proposed approach. We observe that Fuzzy Logic allows us to introduce more flexible data agglomeration techniques. From the analysis of scientific literature, it appears to be the first time that a model based on fuzzy logic is used for the agglomeration of multi-omic data. The results suggest that FH-Clust provides better prognostic value and clinical significance compared to the analysis of single-omic data alone and it is very competitive with respect to other techniques from literature.

Incremental fuzzy clustering for bipolar disorder detection

Gabriella Casalino, Università degli Studi di Bari Aldo Moro

Voice features from everyday phone conversations are regarded as a sensitive digital marker of mood phases in bipolar disorder. At the same time, although acoustic data collected from smartphones are relatively large, their psychiatric labelling is usually very limited, and there is still a need for intelligent and interpretable approaches to process such multiple data streams with a low percentage of labelling. Furthermore, both acoustic data and psychiatric labels are subject to several sources of uncertainty (e.g., irregular phone usage, background noises, subjectivity in psychiatric evaluation). To cope with these characteristics of an acoustic data stream, we propose the use of the Dynamic Incremental Semi-Supervised Fuzzy C-Means algorithm (DISSFCM) for supporting bipolar disorder monitoring. Moreover, we exploit the cluster prototypes from fuzzy semi-supervised learning to analyze the dynamics of data streams and to explain the results. The work illustrates results obtained from real-life data collected from smartphones and psychiatric assessments of a bipolar disorder patient. This is a joint work with the Polish Academy of Sciences, Warsaw, Poland, and it is partially funded from the Italian Ministry of Education, University and Research through the European PON project AIM (Attraction and International Mobility), nr. 1852414, activity 2, line 1.

A combination of genetic algorithms and fuzzy rules for gene expression description

Arianna Consiglio, Istituto di Tecnologie Biomediche CNR - Bari

The advent of Next-Generation Sequencing technologies has revolutionized the study of DNA and RNA in biomedicine. The opportunity to investigate a whole genome (billions of nucleotides) or transcriptome (thousands of genes) with a single biotechnological assay has opened new experimental horizons and many computational issues. The study of the gene expression profiles (number of RNA copies active in cells) of biological samples has allowed to understand many molecular and cellular functions in fisiological and pathological conditions.

The analysis of gene expression data is a complex task, and many tools and pipelines are available to handle big sequencing datasets for case-control (bivariate) studies. In some cases, such as pilot or exploratory studies, the researcher needs to compare more than two groups of samples consisting of a few replicates. Both standard statistical bioinformatic pipelines and innovative deep learning models are unsuitable for extracting interpretable patterns and information from such datasets.

In this presentation, a combination of fuzzy rule systems and genetic algorithms is proposed for the analysis and description of gene expression variations in multiclass experiments. Genetic algorithms are used for the selection of genes and fuzzy rule-based systems for the classification task.

The method has been tested on an ovarian cancer dataset that contains the transcriptome of 21 human ovarian tissue samples from 12 cancer and 9 non-cancer samples, grouped into 6 diagnostic classes [1]. Due to the large number of classes and the low number of replicates for each class, this dataset is quite difficult to analyze with standard bioinformatic tools. After testing several parameters, the final model consists of 10 genes involved in the molecular pathways of cancer and 10 rules that correctly classify all samples.

[1] Consiglio, A., Casalino, G., Castellano, G., Grillo, G., Perlino, E., Vessio, G., Licciulli, F. (2021). Explaining Ovarian Cancer Gene Expression Profiles with Fuzzy Rules and Genetic Algorithms. Electronics, 10(4), 375.

A smart mirror based on fuzzy rules for cardio-vascular risk assessment (Demo)

Gianluca Zaza, Università degli Studi di Bari Aldo Moro

We present MIRACLE, a smart mirror that detects vital parameters by processing the image of the reflected face via the use of remote photoplethysmography techniques. The measured parameters are analyzed by an intelligent component that predicts the cardiovascular risk level using diagnostic rules represented and elaborated through the fuzzy logic formalism. MIRACLE is an economic and non-invasive Personal Healthcare device designed to improve the quality of life of patients suffering from cardiovascular diseases and prevent the occurrence of diseases in healthy patients.

Page updated

Google Sites

Report abuse