US 11,887,700 B2
Techniques for generating encoded representations of compounds
Steven J. Ciavarini, Natick, MA (US); Curt Devlin, Fairhaven, MA (US); Patrick Brophy, Newton Highlands, MA (US); and Scott J. Geromanos, Middletown, NJ (US)
Assigned to WATERS TECHNOLOGIES IRELAND LIMITED, Dublin (IE)
Filed by WATERS TECHNOLOGIES IRELAND LIMITED, Dublin (IE)
Filed on May 4, 2020, as Appl. No. 16/865,880.
Claims priority of provisional application 62/842,694, filed on May 3, 2019.
Prior Publication US 2020/0350039 A1, Nov. 5, 2020
Int. Cl. G16C 20/20 (2019.01); G16C 20/80 (2019.01); G01N 30/00 (2006.01)
CPC G16C 20/20 (2019.02) [G01N 30/00 (2013.01); G16C 20/80 (2019.02)] 18 Claims
OG exemplary drawing
 
1. An apparatus, comprising:
at least one memory; and
logic coupled to the at least one memory, the logic to:
receive analytical information associated with at least one compound,
generate at least one encoded representation of the at least one compound, the encoded representation comprising at least one segment encoding at least one property of the at least one compound using a plurality of symbols, wherein the at least one property comprises one or more of a charge, a mass, an intensity value for a mass-to-charge ratio, an intensity value for a retention time, an intensity value for a drift time, or collision cross-section data, the generating comprising:
accessing a plurality of ion signatures, each ion signature comprising a plurality of bits representing a precursor ion encoding and a plurality of bits representing a product ion encoding,
matching a subset of the plurality of ion signatures based on their respective precursor ion encodings to generate an initial match set,
correlating the ion signatures of the initial match set based on their respective product ion encodings,
retaining, from among the correlated ion signatures, bits illustrating a match rate of a predetermined statistical significance, the at least one encoded representation comprising the retained bits, and
adding the at least one encoded representation to an encoded molecule library configured to provide automated high-throughput screening by providing the at least one encoded representation as a target to identify or quantify molecular species in a sample.