Toward insights on determining factors for high activity in antimicrobial peptides via machine learning

Abstract

The continued and general rise of antibiotic resistance in pathogenic microbes is a well-recognized global threat. Host defense peptides (HDPs), a component of the innate immune system have demonstrated promising potential to become a next generation antibiotic effective against a plethora of pathogens. While the effectiveness of antimicrobial HDPs has been extensively demonstrated in experimental studies, theoretical insights on the mechanism by which these peptides function is comparably limited. In particular, experimental studies of AMP mechanisms are limited in the number of different peptides investigated and the type of peptide parameters considered. This study makes use of the random forest algorithm for classifying the antimicrobial activity as well for identifying molecular descriptors underpinning the antimicrobial activity of investigated peptides. Subsequent manual interpretation of the identified important descriptors revealed that polarity-solubility are necessary for the membrane lytic antimicrobial activity of HDPs.

Comparative transcriptomic and physiological analyses of weedy rice and cultivated rice to identify vital differentially expressed genes and pathways regulating the ABA response

“…5 c). A similar hormone regulatory network was established in wheat roots after ABA treatment 15 , indicating that phytohormone signaling transduction pathways are critical in stress responses.…”

“…Abscisic acid (ABA) is generally considered a stress hormone, and its biosynthesis plays a role in rapid metabolic adjustment in response to stress 14 . The expression of stress-inducible genes in plants is primarily regulated by both ABA-dependent and ABA-independent pathways 15 . When plants encounter stress conditions, endogenous ABA rapidly accumulates.…”

Physicochemical Features and Peculiarities of Interaction of AMP with the Membrane

“…In these cases, design has been coupled with structure-activity relationship studies. Another approach that has been used is an in silico study, relying on machine learning on the sets of data on experimentally explored peptides [ 179 , 180 , 181 , 182 , 183 ]. The aim is to uncover the features of peptides that are most informative in distinguishing one type of peptides from another and based on which an efficient predictive model can be established.…”

“…There are peptides with a high value of total positive charge and peptides with zero or even negative charges; peptides with high hydrophobic moment and peptides which are not amphipathic ( Figure 9 ); peptides with the isoelectric point around the value of 14 and peptides with a lower isoelectric point ( Figure 10 ). Therefore, it is reasonable to think about the grouping of AMPs according to physicochemical features and to assign a particular mode of action for each group of peptides [ 180 ]. A clusterization of peptides according to physicochemical features has recently been performed [ 193 , 194 ].…”

“…Here the question arises: what is an optimal set of physicochemical descriptors that should be used? The authors of [ 174 , 180 ] have used many different descriptors to classify AMP such as charge, hydrophobicity, hydrophobic moment, a propensity to disordered structure, etc. We have to emphasize that proper attention has not been paid to very important physical interactions, such as the cation–π interactions.…”