Methods, systems, and computer readable media for aptamer selection

Patent metadata

Field	Value
Publication number	US-2025279163-A1
Application number	US-202218701423-A
Country	US
Kind code	A1
Filing date	Oct 14, 2022
Priority date	Oct 15, 2021
Publication date	Sep 4, 2025
Grant date	—

Abstract

Provided herein are methods of generating a trained classifier at least partially using a computer. The methods include training a Restricted Boltzmann Machine (RBM) using at least a first training dataset that comprises sequence information corresponding to a population of aptamers, and/or one or more descriptors thereof, which aptamers comprise a minimum threshold binding affinity to a target biomolecule to produce a trained RBM model. Additional methods as well as related systems and computer readable media are also provided.

First claim

What is claimed is:

1 . A method of generating a trained classifier at least partially using a computer, the method comprising training, by the computer, a Restricted Boltzmann Machine (RBM) using at least a first training dataset that comprises sequence information corresponding to a population of aptamers, and/or one or more descriptors thereof, which aptamers comprise a minimum threshold binding affinity to a target biomolecule to produce a trained RBM model, thereby generating the trained classifier at least partially using the computer.

2 . The method of claim 1 , comprising applying a maximum likelihood algorithm to the first training dataset to identify and exclude erroneous information.

3 . The method of claim 1 , wherein the first training dataset comprises one or more sequence motifs.

4 . The method of claim 1 , comprising repeating the training step using at least a second training dataset.

5 . The method of claim 1 , wherein the target biomolecule comprises thrombin.

6 . The method of claim 1 , comprising generating candidate aptamer sequence information using the trained RBM model.

7 . The method of claim 1 , comprising synthesizing the candidate aptamer using the candidate aptamer sequence information to produce a synthesized candidate aptamer.

8 . The method of claim 7 , comprising using the synthesized candidate aptamer to bind the target biomolecule.

9 . The trained RBM model produced by the method of claim 1 .

10 . A method of generating a candidate aptamer, the method comprising:
generating candidate aptamer sequence information using a trained Restricted Boltzmann Machine (RBM) model produced using at least a first training dataset that comprises sequence information corresponding to a population of aptamers, and/or one or more descriptors thereof, which aptamers comprise a minimum threshold binding affinity to a target biomolecule; and, synthesizing the candidate aptamer using the candidate aptamer sequence information, thereby generating the candidate aptamer.

11 . The method of claim 10 , wherein the trained RBM model is generated at least in part by applying a maximum likelihood algorithm to the first training dataset to identify and exclude erroneous information.

12 . The method of claim 10 , wherein the first training dataset comprises one or more sequence motifs.

13 . The method of claim 10 , wherein the target biomolecule comprises thrombin.

14 . The method of claim 10 , comprising using the synthesized candidate aptamer to bind the target biomolecule.

15 . A system, comprising a controller comprising, or capable of accessing, computer readable media comprising non-transitory computer executable instructions which, when executed by at least one electronic processor, perform at least training a Restricted Boltzmann Machine (RBM) using at least a first training dataset that comprises sequence information corresponding to a population of aptamers, and/or one or more descriptors thereof, which aptamers comprise a minimum threshold binding affinity to a target biomolecule to produce a trained RBM model.

16 . The system of claim 15 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: applying a maximum likelihood algorithm to the first training dataset to identify and exclude erroneous information.

17 . The system of claim 15 , wherein the first training dataset comprises one or more sequence motifs.

18 . The system of claim 15 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: repeating the training step using at least a second training dataset.

19 . The system of claim 15 , wherein the target biomolecule comprises thrombin.

20 . The system of claim 15 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: generating candidate aptamer sequence information using the trained RBM model.

21 . The system of claim 15 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: synthesizing the candidate aptamer using the candidate aptamer sequence information and an operably connected a biomolecule synthesis device to produce a synthesized candidate aptamer.

22 . A system, comprising:
a biomolecule synthesis device; and at least one controller operably connected to the biomolecule synthesis device, which controller comprises, or is capable of accessing, computer readable media comprising non-transitory computer executable instructions which, when executed by at least one electronic processor, perform at least:
generating candidate aptamer sequence information using a trained Restricted Boltzmann Machine (RBM) model produced using at least a first training dataset that comprises sequence information corresponding to a population of aptamers, and/or one or more descriptors thereof, which aptamers comprise a minimum threshold binding affinity to a target biomolecule; and,
synthesizing the candidate aptamer using the biomolecule synthesis device and the candidate aptamer sequence information.

23 . The system of claim 22 , wherein the trained RBM model is generated at least in part by applying a maximum likelihood algorithm to the first training dataset to identify and exclude erroneous information.

24 . The system of claim 22 , wherein the first training dataset comprises one or more sequence motifs.

25 . The system of claim 22 , wherein the target biomolecule comprises thrombin.

26 . A computer readable media comprising non-transitory computer executable instruction which, when executed by at least electronic processor perform at least training a Restricted Boltzmann Machine (RBM) using at least a first training dataset that comprises sequence information corresponding to a population of aptamers, and/or one or more descriptors thereof, which aptamers comprise a minimum threshold binding affinity to a target biomolecule to produce a trained RBM model.

27 . The computer readable media of claim 26 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: applying a maximum likelihood algorithm to the first training dataset to identify and exclude erroneous information.

28 . The computer readable media of claim 26 , wherein the first training dataset comprises one or more sequence motifs.

29 . The computer readable media of claim 26 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: repeating the training step using at least a second training dataset.

30 . The computer readable media of claim 26 , wherein the target biomolecule comprises thrombin.

31 . The computer readable media of claim 26 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: generating candidate aptamer sequence information using the trained RBM model.

32 . The computer readable media of claim 26 , wherein the executable instructions which, when executed by the electronic processor, further perform at least: synthesizing the candidate aptamer using the candidate aptamer sequence information and an operably connected a biomolecule synth

Assignees

• Univ Arizona State
• Centre Nat Rech Scient
• Ecole Normale Superieure
• Univ Sorbonne
• Univ Paris Cite

Inventors

• Sulc Petr
• Monasson Remi
• Cocco Simona

Classifications

• C12N2320/13

  in a process of directed evolution, e.g. SELEX, acquiring a new function · CPC title

• C12N2310/16

  Aptamers · CPC title

• C12N15/115 Primary

  Aptamers, i.e. nucleic acids binding a target molecule specifically and with high affinity without hybridising therewith {; Nucleic acids binding to non-nucleic acids, e.g. aptamers} · CPC title

• C12N15/1048

  SELEX · CPC title

• G06N3/047

  Probabilistic or stochastic networks · CPC title