Metagenomic library and natural product discovery platform

Abstract

The present disclosure provides methods and systems for identifying natural product-encoding multi-gene clusters (MGCs). In some embodiments, the present disclosure also teaches methods for producing sequenced and assembled metagenomic libraries that are amenable to MGC search bionformatic tools and techniques.

1 . An in silico method for identifying a candidate multi-gene cluster (MGC) that does not encode for a known resistance gene, said method comprising the steps of:
 a) providing the sequence of a known or predicted MGC;   b) computationally predicting natural product multi-gene cluster feature sets within a long-assembly digital metagenomic library and supplying the output of said prediction as a plurality of signal-associated multi-gene cluster digital feature sets;   c) selecting a candidate MGC from amongst the plurality of signal-associated multigene cluster digital feature sets of step (b), said candidate MGC comprising at least one similarity factor selected from the group consisting of:
 i) sequence homology of 1, 2, 3, 4, 5, 6, 7, or 8 biosynthetic enzymes within the known or predicted MGC and the candidate MGC; 
 ii) same number of each type of biosynthetic module(s) within the known or predicted MGC and the candidate MGC; and 
 iii) similarity of the predicted chemical structures of natural products produced by the known/predicted MGC and the candidate MGC; 
   
       thereby identifying the candidate MGC that does not encode for a known resistance gene. 
     
     
         2 . A method for biosynthetic analoging of a target natural product, said method comprising the steps of:
 a) providing a plurality of genetic sequences, each encoding an enzyme known or predicted to catalyze a type of reaction for a first analoging of the target natural product;   b) perturbing the genome of one or more cells of a first base microbial strain to each express an enzyme encoded by one or more of the plurality of genetic sequences of step (a), wherein the first base microbial strain is capable of synthesizing the target natural product, thereby creating an analoging enzyme panel library of microbial strains;   c) culturing individual microbial strains from the analoging enzyme panel library of microbial strains;   d) analyzing spent media from the cultures of step (c), for the presence of the target natural product and/or analogs of said target natural product; and   e) selecting a microbial strain from the analoging enzyme panel of microbial strains, wherein the selected microbial strain produces a desired analog of the target natural product, as determined by the analysis of step (d), thereby analoging the target natural product.   
     
     
         3 . An in silico method for searching a digital metagenomics library and identifying a natural product of interest, comprising:
 a) querying a digital metagenomics library for a signal indicative of a natural product multi-gene cluster feature set;   b) supplying the output of said query as a plurality of signal-associated multi-gene cluster digital feature sets;   c) determining and assigning biologic relevancy to the signal-associated multi-gene cluster digital feature sets, by:   determining a computationally predicted biosynthetic functionality of a plurality of genes from the signal-associated multi-gene cluster digital feature set and digitally assembling a computationally determined natural product multi-gene cluster (MGC) feature set comprising one or more biosynthetic operon(s); and/or   determining a computationally predicted biological resistance gene functionality of at least one gene from a signal-associated multi-gene cluster digital feature set to thereby identify a computationally determined biological resistance gene; and   
       d) identifying an MGC encoding for the natural product of interest based upon a computationally determined biological resistance gene being located within a threshold parameter of a computationally determined natural product multi-gene cluster feature set comprising a biosynthetic operon; 
       wherein the digital metagenomics library comprises an N50 length of at least about 15 kb, and is at least about 500 MB in size.