Apparatus and method for extended depth of field

Patent metadata

Field	Value
Publication number	US-12449362-B2
Application number	US-202418879735-A
Country	US
Kind code	B2
Filing date	Mar 14, 2024
Priority date	Mar 17, 2023
Publication date	Oct 21, 2025
Grant date	Oct 21, 2025

Abstract

A method for extending a depth of field of a nucleic acid sequencer may comprise optimization steps which are repeated one or more times, in which a result of passing light through an objective lens and a mask is compared with an ideal result, and any discrepancy is used to update the mask. Such a mask may be incorporated into a nucleic acid sequencer which adds fluorescent tags to nucleic acid sites and then detect light emitted from the fluorescent tags, thereby extending the sequencer's depth of field.

First claim

What is claimed is:

1. A method for extending a depth of field of a nucleic acid sequencer, the method comprising:
performing a set of optimization steps comprising:
determining a first result, wherein the first result is a result of passing light emitted by a sample at a nucleic acid site through an objective lens and a mask, wherein the objective lens is focused on a plane which is separated from the nucleic acid site by a target defocus amount, and wherein the mask is disposed between the objective lens and a set of detection pixels;
determining whether there is a discrepancy between the first result and a second result, wherein the second result is an ideal result for detecting light emitted by the sample at the nucleic acid site; and
based on determining there is the discrepancy between the first result and the second result, performing a set of updating tasks comprising updating the mask based on the determined discrepancy between the first result and the second result; and

repeating the set of optimization steps one or more times.

2. The method of claim 1 , wherein:
the first result is a first point spread function (PSF) indicating how light emitted by the sample at the nucleic acid site is detected by the set of detection pixels after passing through the objective lens and the mask when the objective lens is out of focus by a target defocus amount; and
the second result is a second PSF indicating how light emitted by the sample at the nucleic acid site is detected by the set of detection pixels when the sample at the nucleic acid site is focused on a surface of the sample.

3. The method of claim 2 , wherein:
the nucleic acid sequencer is configured to detect light signals emitted from multiple surfaces of a multi-surface flowcell;
updating the mask based on the determined discrepancy comprises applying a loss function, wherein the piecewise loss function comprises:
a first loss component configured to ensure that signals from a first surface of the multi-surface flowcell are in focus; and
a second loss component configured to ensure that signals from a second surface of the multi-surface flowcell are out of focus.

4. The method of claim 1 , wherein:
the mask is an amplitude mask;
the sample at the nucleic acid site comprises a plurality of clusters;
the set of optimization steps comprises:
separating the plurality of clusters into a first set of clusters and a second set of clusters; and
updating the mask based on the determined discrepancy comprises determining a loss based on a mean and a standard deviation of the first set of clusters, and a mean and a standard deviation of the second set of clusters.

5. The method of claim 1 , wherein:
the first result is a sequence of base calls generated by a base calling algorithm based on light emitted from the sample at the nucleic acid site and detected by the set of detection pixels after passing through objective lens and the mask; and
the second result is a sequence of base calls for a reference genome.

6. The method of claim 5 , wherein:
the base calling algorithm comprises a machine learning model; and
the set of updating tasks comprises updating the machine learning model based on the discrepancy between the first result and the second result.

7. The method of claim 1 , wherein:
the nucleic acid sequencer is configured to detect light signals in a plurality of wavelengths;
the first result is a result of passing a first wavelength of light emitted by the sample at the nucleic acid site through the objective lens and the mask; and
the set of optimization steps comprises:
determining a third result, wherein the third result is a result of passing a second wavelength of light emitted by the sample at the nucleic acid site through the objective lens and a second mask, wherein the objective lens is focused on the plane which is separated from the nucleic acid site by the target defocus amount, and wherein the second mask is disposed between the objective lens and a second set of detection pixels;
determining whether there is a discrepancy between the second result with and the third result; and
based on determining there is a discrepancy between the second result and the third result, performing a second set of updating tasks comprising updating the second mask based on the determined discrepancy between the second result and the third result.

8. The method of claim 1 , wherein the mask is a phase mask.

9. The method of claim 1 , wherein, on each repetition of the set of optimization steps:
determining the first result is performed a plurality of times, wherein each time from the plurality of times, the target defocus amount is different from the target defocus amount on each other time from the plurality of times;
performing the set of updating tasks is performed a set of times, wherein each time the set of updating tasks is performed, a loss is calculated which corresponds to the target defocus amount from a different time from the plurality of times.

10. The method of claim 9 , wherein, on each repetition of the set of optimization steps, the plurality of times the first result is determined has a cardinality less than the set of times the set of updating tasks is performed.

11. The method of claim 10 , wherein, on each repetition of the set of optimization tasks:
the set of updating tasks comprises determining the target defocus resulting in the largest discrepancies between the first result and the second result; and
each loss which is calculated during performance of the set of updating tasks corresponds to a target defocus amount which results in a greater discrepancy between the first result and the second result than any target defocus amount which does not correspond to a loss calculated during that repetition of the set of optimization tasks.

12. The method of claim 1 , wherein set of optimization steps are performed in simulation.

Assignees

• Illumina Inc

Inventors

• Gau Jeffrey
• Pinto Joseph
• Dutta Anindita
• Evans Geraint
• Wang Yina
• Prince Simon
• Hong Stanley

Classifications

• G01N2021/6439

  with indicators, stains, dyes, tags, labels, marks · CPC title

• G01N2021/6441

  with two or more labels · CPC title

• G02B21/16

  adapted for ultraviolet illumination {; Fluorescence microscopes (G02B21/0076 takes precedence)} · CPC title

• G02B27/0075

  with means for altering, e.g. increasing, the depth of field or depth of focus · CPC title

• G01N21/6458

  Fluorescence microscopy (fluorescence microscopes per se G02B21/0076 and G02B21/16) · CPC title