US 12,442,035 B2
Compositions and methods for detecting antibiotic responsive mRNA expression signatures and uses thereof
Deborah Hung, Cambridge, MA (US); Roby Bhattacharyya, Boston, MA (US); Jonathan Livny, Cambridge, MA (US); and Peijun Ma, Cambridge, MA (US)
Assigned to THE BROAD INSTITUTE, INC., Cambridge, MA (US); and THE GENERAL HOSPITAL CORPORATION, Boston, MA (US)
Appl. No. 17/271,496
Filed by THE BROAD INSTITUTE, INC., Cambridge, MA (US); and THE GENERAL HOSPITAL CORPORATION, Boston, MA (US)
PCT Filed Aug. 26, 2019, PCT No. PCT/US2019/048114
§ 371(c)(1), (2) Date Feb. 25, 2021,
PCT Pub. No. WO2020/046801, PCT Pub. Date Mar. 5, 2020.
Claims priority of provisional application 62/834,786, filed on Apr. 16, 2019.
Claims priority of provisional application 62/723,417, filed on Aug. 27, 2018.
Prior Publication US 2021/0230675 A1, Jul. 29, 2021
Int. Cl. C12Q 1/6816 (2018.01); C12N 1/06 (2006.01); C12N 1/20 (2006.01); C12Q 1/6876 (2018.01); C12Q 1/689 (2018.01)
CPC C12Q 1/6816 (2013.01) [C12N 1/06 (2013.01); C12N 1/20 (2013.01); C12Q 1/6876 (2013.01); C12Q 1/689 (2013.01)] 11 Claims
 
1. A method, comprising:
obtaining a sample including one or more bacterial cells, wherein the sample is obtained from a patient;
processing the sample to enrich the one or more bacterial cells;
contacting the sample with one or more antibiotic compounds;
lysing the sample to release messenger ribonucleic acid (mRNA) from the one or more bacterial cells;
hybridizing the released mRNA to at least one set of two nucleic acid probes, wherein each nucleic acid probe includes a unique barcode or tag and wherein the at least one set of nucleic acid probes includes one or more probes from SEQ ID NOs: 74-86 and one or more probes from SEQ ID NOs: 91-103;
detecting the hybridized nucleic acid probes;
identifying one or more genetic resistance determinants;
determining antibiotic susceptibility of the one or more bacterial cells using a machine learning algorithm trained on a susceptibility classification dataset, wherein the machine learning algorithm generates a ranked list of two or more susceptibility features, thereby determining antibiotic susceptibility of the one or more bacterial cells; and
selecting an appropriate antibiotic based on the determined antibiotic susceptibility of the one or more bacterial cells; and
administering the appropriate antibiotic to the patient.