StaLAENet: A stacked LSTM-nested deep-autoencoder network for identification of antimicrobial resistance of nosocomial pathogens.

As various technological innovations are assisting medical science in a considerable way, rendering a significant leap towards 'lab-to-land' delivery, in a similar vein, algorithm development and concomitant framework-based approaches help the field to enrich its patient care. Although antimicrobial drugs revolutionized this particular area, antimicrobial resistance is a pressing global health concern as microbial strains are becoming resistant to conventional antibiotics, undermining the efficacy of these drugs and leading to increased illness and healthcare costs. To tackle this menace, apart from technological innovations such as diagnostic kits, an informatics-based framework approach is the call of the day. Despite the emergence of several computational approaches, they lack in generalization, scope, and scalability. Here, we have developed a novel framework StaLAENet (stacked LSTM-nested deep-autoencoder network) to predict antibiotic-resistant gene drug classes targeting ESKAPE pathogens. This framework comprises two modules: a feature representation module comprising a stacked LSTM-nested deep autoencoder and a classification module that leverages a dense network using latent features. StaLAENet demonstrated an efficient performance - accuracy: 0.938±0.043, specificity: 0.888±0.061, precision: 0.912±0.020, and recall: 0.881±0.021 - for