International group of bioinformatics scientists has developed Dereplicator, a program which helps to identify waste products of bacteria, i.e. potential antibiotics, anticancer and antiviral drugs. Junior research fellows Alexey Gurevich and Alla Mikheenko actively participated in designing and improving the software, which was partly developed in the laboratory “Center for Algorithmic Biotechnology” at SPbU, headed by Prof. Pavel Pevzner.
Dereplicator allows scientists to compare and assess the databases of the chemical structures of the waste products of microorganisms, so called natural products, and input data in form of mass-spectra of the necessary substances. The software identifies which chemical compounds from the input data have already been included in the databases of known natural products. Thus, ignoring the replicas which have already been studied and registered in the databases facilitates the search and helps to focus on those substances which can potentially be new antibiotics.
Dereplicator is specially designed for working with natural products, i.e. compounds which are produced by bacteria and more complex organisms in the process for survival. These substances have great potential to be natural antibiotics. Thus, they are of great interest for scientists as developing new anti-infective drugs is a highly important task nowadays. The natural products are not an easy nut to crack in science as they consist of non-standard amino acids, apart from 20 standard amino acids contained in proteins. These non-standard amino acids are formed into complex chemical structures. Unlike its predecessors, the software is a breakthrough since it can analyse more complex cyclic and branching structures, rather than just regular proteins, a linear sequence of the standard amino acids.
SPbU scientists have successfully improved the software and made it usable by scientific community worldwide. To that end, they developed a web-version of the software, which is available to the users registered at the GNPS (The Global Natural Product Social Molecular Networking (http://gnps.ucsd.edu/). The software was launched at GNPS platform in December 2015, which allows chemists and biologists to use the software through an intuitive web-interface. Before, the use of the software was restricted since it was initially developed as a console application for Linux and Mac OS X.
Moreover, SPbU bioinformaticians significantly sped up the software. In September 2015 it took one week to launch the first version of the program on the whole GNPS publicly available data – today Dereplicator can process the same volume of data in 4 hours. Also, scientists from St. Petersburg added visualization of the tool’s output: now it can draw chemical structures of the found compounds and their correspondence to mass-spectrum peaks; fixed a number of software errors which decreased accuracy and implemented a function to calculate false discovery rate.
“We have managed to make Dereplicator accessible and intuitive, – said Alexey Gurevich. – The program is easy to launch, it is fast and presents results in visual forms. Obviously, we are going to further improve the tool to help writing essay scientists in their endeavours to discover new antibiotics, anticancer and antiviral drugs more efficiently”.
Learn more: Dereplication of peptidic natural products through database search of mass spectra, Nature Chemical Biology, October 31, 2016.
*Mass-spectrum is a mass-to-charge ratio.