Classifying diseases by using biological features to identify potential nosological models

Abstract

Established nosological models have provided physicians an adequate enough classification of diseases so far. Such systems are important to correctly identify diseases and treat them successfully. However, these taxonomies tend to be based on phenotypical observations, lacking a molecular or biological foundation. Therefore, there is an urgent need to modernize them in order to include the heterogeneous information that is produced in the present, as could be genomic, proteomic, transcriptomic and metabolic data, leading this way to more comprehensive and robust structures. For that purpose, we have developed an extensive methodology to analyse the possibilities when it comes to generate new nosological models from biological features. Different datasets of diseases have been considered, and distinct features related to diseases, namely genes, proteins, metabolic pathways and genetical variants, have been represented as binary and numerical vectors. From those vectors, diseases distances have been computed on the basis of several metrics. Clustering algorithms have been implemented to group diseases, generating different models, each of them corresponding to the distinct combinations of the previous parameters. They have been evaluated by means of intrinsic metrics, proving that some of them are highly suitable to cover new nosologies. One of the clustering configurations has been deeply analysed, demonstrating its quality and validity in the research context, and further biological interpretations have been made. Such model was particularly generated by OPTICS clustering algorithm, by studying the distance between diseases based on gene sharedness and following cosine index metric. 729 clusters were formed in this model, which obtained a Silhouette coefficient of 0.43.

Combined Unsupervised Machine Learning and clinical cohort for phenotyping complex diseases with its application to Amyotrophic lateral sclerosis

“…The index is the ratio of the sum of between-clusters dispersion and within-cluster dispersion for all clusters (where dispersion is de ned as the sum of distances squared). The sum of square errors (SSE), also known as 'inertia' or 'dispersion', represents the sum of squared distances of samples to their closest cluster center 25 . The smaller the SSE value, the better the clustering effect.…”

Why the way we define diseases prevents innovation and precision medicine

“…This is no longer manageable by a single specialist but requires network bioinformatics to analyse the clustering of different phenotypes due to shared genes, symptoms, drugs, or comorbidity associations 61 . These mechanistic definitions can then be captured in a molecular disease classification system 67 . For most NCDs, this will mean endo-or subtyping of the current organ-or symptom-based umbrella terms 65 .…”

Analysis of Antioxidant Capacity Variation among Thai Holy Basil Cultivars (Ocimum tenuiflorum L.) Using Density-Based Clustering Algorithm

“…Nevertheless, the nonexplicit outputs require further analysis procedures to enable practitioners to determine the cluster patterns [31]. OPTICS has been utilized to cluster the integration of omics data, such as disease features (e.g., genes, proteins, pathways, and variants) [32], larval instars [33], wheat genotypic data [34], and metabolic features [35].…”