A Novel Feature Selection Method for Uncertain Features: An Application to the Prediction of Pro-/Anti-Longevity Genes

Abstract

Understanding the ageing process is a very challenging problem for biologists. To help in this task, there has been a growing use of classification methods (from machine learning) to learn models that predict whether a gene influences the process of ageing or promotes longevity. One type of predictive feature often used for learning such classification models is Protein-Protein Interaction (PPI) features. One important property of PPI features is their uncertainty, i.e., a given feature (PPI annotation) is often associated with a confidence score, which is usually ignored by conventional classification methods. Hence, we propose the Lazy Feature Selection for Uncertain Features (LFSUF) method, which is tailored for coping with the uncertainty in PPI confidence scores. In addition, following the lazy learning paradigm, LFSUF selects features for each instance to be classified, making the feature selection process more flexible. We show that our LFSUF method achieves better predictive accuracy when compared to other feature selection methods that either do not explicitly take PPI confidence scores into account or deal with uncertainty globally rather than using a per-instance approach. Also, we interpret the results of the classification process using the features selected by LFSUF, showing that the number of selected features is significantly reduced, assisting the interpretability of the results. The datasets used in the experiments and the program code of the LFSUF method are freely available on the web at http://github.com/pablonsilva/FSforUncertainFeatureSpaces.

An Ensemble of Naive Bayes Classifiers for Uncertain Categorical Data

“…In this work, we have applied the proposed approach to the classification of ageing-related genes in different types of organisms, which several studies have addressed in recent years [8]- [10], [14]- [16]. In this problem, the objective is to identify the effect of genes on the longevity of an organism.…”

“…Genes encode proteins, and information about the interaction between two proteins, known as protein-protein interaction (PPI), has been used in past work addressing the classification of ageing-related genes [10], [14], [16], leading to improvements in predictive performance. STRING [18] is a database of PPIs that stem from computational predictions, from knowledge transfer between organisms, and from interactions aggregated from other databases.…”

“…We have chosen Naive Bayes as the base classifier to evaluate our proposed ensemble scheme since it has obtained good results in many real-world domains, including the classification of ageing-related genes [8]- [10] -the target application domain in this work. Furthermore, there are several reports on the use of Naive Bayes classifiers to build ensembles in the literature [11]- [13].…”

Feature importance feedback with Deep Q process in ensemble-based metaheuristic feature selection algorithms

“…Embedded methods 11 – 13 such as LASSO and Decision Trees, incorporate feature selection within the algorithm itself for different scenarios. These methods usually provide a good trade-off between performance and speed but are limited by the biases of the algorithms they are integrated with for different use cases 14 – 16 . Genetic Algorithms (GAs) 17 – 19 use principles of natural evolution to explore the feature space.…”