The vast majority of microorganisms on Earth reside in often-inseparable environment-specific communities – microbiomes. Meta-genomic/-transcriptomic sequencing could reveal the otherwise inaccessible functionality of microbiomes. However, existing analysis approaches focus on attributing sequencing reads to known genes/genomes and, thus, often fail to make maximal use of available data.
Metal binding proteins have a central role in metabolism and catalysis. Nearly one third of known protein structures contain metal ions that are used for a variety of critical needs, such as catalysis, DNA/RNA binding, protein structure maintenance, etc. Identifying metal binding proteins is thus crucial for our understanding of the mechanisms of cellular activity. Protein sequences identified from putative open reading frames in sequenced genomes are readily available for analysis. However, the type of information required to identify a metal binding protein is only available for a very limited amount of proteins.
We developed a novel machine learning-based method for identifying metal binding proteins from sequence-derived features. Our method is over 90% accurate around for identifying proteins that bind ten ubiquitously present metal ligands (Fe, Ca, Na, K, Mg, Mn, Cu, K, Co, Ni). When compared to current methods (i.e BLAST, MetalDetector2) MeBiPred shows a considerable improvement on detection capabilities.
Citing MeBiPred If you use MeBiPred in published research, please cite:
Aptekmann AA, Buongiorno J, Giovanelli D, Ferreiro DU, Bromberg Y. (2021). MeBiPred: A powerful tool to discover metal binding proteins.
Read this publication at Nucleic Acids Research (doi:10.1093).
Get the latest release of MeBiPred and run analyses locally
Download the MeBiPred sources ( ) or simply install MeBiPred as python package: pip install mymetal ( )