HELP: A computational framework for labelling and predicting human context-specific essential genes
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Machine learning-based approaches are particularly suitable for identifying essential genes as they allow the generation of predictive models trained on features from multi-source data. Gene essentiality is neither binary nor static but determined by the context. The databases for essential gene annotation do not permit the personalisation of the context, and their update can be slower than the publication of new experimental data. We propose HELP ( H uman Gene E ssentiality L abelling & P rediction), a computational framework for labelling and predicting essential genes. Its double scope allows for identifying genes based on dependency or not on experimental data. The effectiveness of the labelling method was demonstrated by comparing it with other approaches in overlapping the state-of-the-art EG annotations, where HELP demonstrated the best compromise between false and true positive rates. The gene attributes, including multi-omics and network embedding features, lead to high-performance prediction of EGs while confirming the existence of essentiality nuances.
Author summary
Essential genes (EGs) are commonly defined as those required for an organism or cell’s growth and survival. The essentiality is strictly dependent on both environmental and genetic conditions, determining a difference between those considered common EGs (cEGs), essential in most of the contexts considered, and those essential specifically to one or few contexts (context-specific EGs, csEGs). In this paper, we present a library of tools to address the identification and prediction of csEGs. Furthermore, we attempt to experimentally explore the statement that essentiality is not a binary property by identifying, predicting and analyzing an intermediate class between the Essential (E) and Not Essential (NE) genes. Among the multi-source data used to predict the EGs, we found the best attributes combination to capture the essentiality. We demonstrated that the additional class of genes we defined as “almost Essential” shows differences in these attributes from the E and NE genes. We believe that investigating the context-specificity and the dynamism of essentiality is particularly relevant to unravelling crucial insights into biological mechanisms and suggesting new candidates for precision medicine.
