HIGH-THROUGHPUT NOVEL ANTIMICROBIAL DISCOVERY FROM 'OMIC DATASOURCES
AMPLY is a bioprospecting pipeline that can accept a wide variety of ‘omic data sources (including metagenomic, metatranscriptomic, genomic and raw sequencing reads) and applies a variety of techniques (see sidebar) to highlight novel, therapeutically viable bioactive peptides.
These peptides are then bulk synthesised from scratch and screened for activity against a wide variety of clinical pathogens, including antibiotic resistant Gram negative and Gram positive bacteria. Positive and negative in vitro data is loaded directly into the AMPLY software to improve in silico detection rates.
AMPLY moves beyond bioinformatics and leverages peptide synthesis technologies to physically create any compounds identified by the bioinformatic pipeline. With a variety of pricing and purification options projects can generate hundreds of compounds for a wider screening programme, or focus on a smaller subset of compounds designed to meet a specific need.
The output from the lab testing work is a profile for each compound of how bio-efficient it is at killing a suite of target organisms (either clinical, or designated as targets through the the project) and a variety of other metrics, including the toxicity of the compounds and indications on a possible mode of action (how they kill). This information is vital in moving these compounds for publication in scientific journals, or presenting as a saleable product to a pharmaceutical company.
Example testable microorganisms
MDR E. coli
Example in vitro characterisation
TEM visualisation of mode of action
De-noising and clustering
AMPLY removes “noise” from datasets through clustering techniques - locating useful bioactive peptides that might be missed by other annotation pipelines.
AMPLY contains over 21,000 known AMPs from a number of online sources with a wide variety of classification grouping and associated metadata. AMPLY, uniquely also has large negative data sets which are used to remove false positives and target bioactivity.
Hidden Markov Models
AMPLY uses Hidden Markov models to filter and highlight putative AMPs and also to provide familial classification.
Regional bioactivity prediction
AMPLY scans putative peptides to search for bioactive regions, so shorter peptide lengths can be synthesised, or longer peptides can be analysed in smaller sections.
Neural Network modelling
AMPLY uses a wide variety of modelling techniques, including Neural Networks to filter prospects for therapeutic potential (high bioactivity, low haemolytic potential). These models are tuned for high-throughput analysis and have excellent negative and positive data separation rates.
Before synthesis, AMPLY can warn of “out of range” characteristics for any prospects.
AMPLY’s mix of annotation methodologies are presented on a series of detailed spreadsheets allowing a user to assess prospects in a single location. The key output is the “BitPad”, which presents over 200 datapoints combined in a single location, presenting an end user with sufficient data to make informed choices on prospects to investigate further.
Contact us to discuss the data you have and how we might be able to assist in its analysis
A variety of pricing strategies are available, depending on the nature of the investigation
Mining and Synthesis Stage
Once the data is delivered we will manage the analysis, synthesis and testing and provide you with a simple report
Once a project is complete we can assist in taking the compounds further, either for publication or patenting.