Barbosa, M. C. | Search bioRxiv

bioRxiv2024

CODARFE: Unlocking the prediction of continuous environmental variables based on microbiome

Despite the surge in data acquisition, there is a limited availability of tools capable of effectively analyzing microbiome data that identify correlations between taxonomic compositions and continuous environmental factors. Furthermore, existing tools also do not predict the environmental factors in new samples, underscoring the pressing need for innovative solutions to enhance our understanding of microbiome dynamics and fulfill the prediction gap. Here, we introduce CODARFE, a novel tool for sparse compositional microbiome-predictors selection and prediction of continuous environmental factors. We tested CODARFE against four state-of-the-art tools in two experiments. First, CODARFE outperformed predictor selection in 21 out of 24 databases in terms of correlation. Second, among all the tools, CODARFE achieved the highest number of previously identified bacteria linked to environmental factors for human data--that is, at least 7% more. We also tested CODARFE in a cross-study, using the same biome but under different external effects (e.g., ginseng field and cattle for arable soil, and HIV and crohns disease for human gut), using a model trained on one dataset to predict environmental factors on another dataset, achieving 11% of mean absolute percentage error. Finally, CODARFE is available in five formats, including a Windows version with a graphical interface, to installable source code for Linux servers and an embedded Jupyter notebook available at MGnify - https://github.com/alerpaschoal/CODARFE.

bioinformatics↗

bioRxiv2024

Capturing G protein-coupled receptors into native lipid-bilayer nanodiscs using new diisobutylene/maleic acid (DIBMA) copolymers

Many membrane proteins, including G-protein-coupled receptors (GPCRs), are susceptible to denaturation when extracted from their native membrane by detergents. Therefore, alternative methods have been developed, including amphiphilic copolymers that enable the direct extraction of functional membrane proteins along with their surrounding lipids, leading to the formation of native lipid-bilayer nanodiscs. Among these amphiphilic copolymers, styrene/maleic acid (SMA) and diisobutylene/maleic acid (DIBMA) polymers have been extensively studied and successfully utilized to isolate various types of membrane proteins, including GPCRs. Despite their many benefits, SMA and DIBMA polymers also have significant drawbacks that limit their application. Most notably, both SMA and DIBMA carry high negative charge densities, which can interfere with protein-protein and protein-lipid interactions through unspecific Coulombic attraction or repulsion. Herein, we describe a series of new amphiphilic copolymers derived from DIBMA via partial amidation of the carboxylate pendant groups with various biocompatible amines. The nanodisc-forming properties of the new polymers were assessed using model membranes as well as in the context of extracting the melanocortin 4 receptor (MC4R), a prototypical class A GPCR. While each new DIBMA variant displays distinct features that may be favorable for selected applications, we identified a new PEGylated DIBMA variant called mPEG4-DIBMA as particularly promising for the studied purpose. On the one hand, mPEG4-DIBMA abolishes unspecific interactions with the tested peptide ligand, a prerequisite for reliably characterizing GPCR-ligand interactions. On the other hand, mPEG4-DIBMA outperforms other polymers such as SMA and DIBMA by achieving higher extraction efficiencies of MC4R from Sf9 insect cell membranes. Thus, this new nanodisc-forming polymer combines two key advantages that are crucial for investigating GPCRs in a well-defined but still native lipid-bilayer environment, thus paving the way for manifold future applications.

biochemistry↗