Methods and systems for digitally counting features on arrays

• US 9,905,005 B2
• Filed: 01/18/2017
• Issued: 02/27/2018
• Est. Priority Date: 10/07/2013
• Status: Active Grant

• Current Assignee: Becton, Dickinson & Co
• Original Assignee: Cellular Research, Inc. (Becton, Dickinson & Co)
• Inventors: Fu, Glenn, Rudoff, Roger, Stern, David, Wu, George T., Fodor, Stephen P. A., Facer, Geoffrey
• Primary Examiner(s): Liew, Alex
• Application Number:
  US15/409,355
• Publication Number:
  US 20170154421A1
• Time in Patent Office: 405 Days
• Field of Search: 382128-134
• CPC Class Codes: G01N 2021/6439 with indicators, stains, dyes, tags, labels, marks
• CPC Class Codes: G01N 21/274 Calibration, base line adjustment, drift correction
• CPC Class Codes: G01N 21/6428 Measuring fluorescence of fluorescent products of reactions or of fluorochrome labelled reactive substances, e.g. measuring quenching effects, using measuring "optrodes" in vivo A61B5/00; immunoassay G01N33/53
• CPC Class Codes: G01N 21/6452 Individual samples arranged in a regular 2D-array, e.g. multiwell plates
• CPC Class Codes: G01N 21/6456 Spatial resolved fluorescence measurements; Imaging
• CPC Class Codes: G01N 2201/129 Using chemometrical methods
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06T 2207/10056 Microscopic image
• CPC Class Codes: G06T 2207/10064 Fluorescence image
• CPC Class Codes: G06T 2207/30072 Microarray; Biochip, DNA array; Well plate
• CPC Class Codes: G06T 2207/30242 Counting objects in image
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/277 involving stochastic approaches, e.g. using Kalman filters

Abstract

Methods, systems and platforms for digital imaging of multiple regions of an array, and detection and counting of the labeled features thereon, are described.

Claims

1. A method for determining the number of a target molecule on an array, comprising:

(a) acquiring image data of a region of an array, wherein the array comprises a plurality of stochastically labeled and non-labeled features, and wherein the features comprise oligonucleotide probes;

(b) measuring a signal intensity and a local background intensity for one of the plurality of stochastically labeled and non-labeled features based on the acquired image data;

(c) calculating a local background corrected signal intensity for the one of the plurality of stochastically labeled and non-labeled features using the signal intensity and the local background intensity measured in (b);

(d) transforming the local background corrected signal intensity calculated in (c) for the one of the plurality of stochastically labeled and non-labeled features to determine the number of stochastically labeled and non-labeled features in the of the array; and

(e) calculating the number of a target molecule based on the number of stochastically labeled and non-labeled features determined in (d) within the region of the array.

2. The method of claim 1, wherein acquiring the image data of the region of the array comprises acquiring the image data of the region of the array from an optical imaging system, wherein the optical imaging system is a fluorescence imaging system or a phosphorescence imaging system, and wherein the optical imaging system operates in a transmitted light, reflected light, or scattered light imaging mode, or any combination thereof.

3. The method of claim 1, wherein the oligonucleotide probes comprise a fluorescein dye, a Cy3 dye, a Cy5 dye, or a phycoerythrin fluorophore.

4. The method of claim 1, further comprising illuminating the region of the array using an illumination system prior to (a) acquiring the image data of the region of the array.

5. The method of claim 1, wherein acquiring the image data of the region of the array comprises automatically locating one or more of the plurality of stochastically labeled and non-labeled features of the array.

6. The method of claim 1, further comprising analyzing the local background corrected signal intensities for the plurality of stochastically labeled and non-labeled features to determine a dynamic signal intensity threshold for distinguishing the stochastically labeled and non-labeled features of the array.

7. The method of claim 6, wherein determining the dynamic signal intensity threshold for distinguishing the stochastically labeled and non-labeled features of the array comprises performing, a k-means clustering analysis, a k-medoids clustering analysis, a mixture model statistical analysis, an empirical analysis based on sorting of the background corrected signal intensity values for the plurality of stochastically labeled and non-labeled features, or an empirical analysis based on sorting of pairwise differences in background corrected signal intensity values for the plurality of stochastically labeled and non-labeled features to determine a dynamic signal intensity threshold for distinguishing the plurality of stochastically labeled and non-labeled features of the array.

8. The method of claim 6, wherein determining the dynamic signal intensity threshold for distinguishing the stochastically labeled and non-labeled features of the array comprises maximizing a quality metric relating to a statistical difference between feature intensities above the dynamic signal intensity threshold and feature intensities below the dynamic signal intensity threshold.

9. The method of claim 6, wherein calculating the local background corrected signal intensity for the one of the plurality of stochastically labeled and non-labeled features using the signal intensity and the local background intensity measured in (b) comprises:

(i) centering a predefined analysis window on an array feature within an image;

(ii) calculating an intensity value statistic for signal and background pixels according to a predefined pattern of pixels within the array feature; and

(iii) utilizing the intensity value statistic for signal and background pixels to calculate the local background corrected signal intensity for the array feature.

10. The method of claim 1, wherein the intensity value statistic used for calculating the background corrected signal intensity for each of the plurality of stochastically labeled and non-labeled features is the mean, the median, or the ratio of signal to background intensities.

11. The method of claim 1, wherein calculating the number of the target molecule based on the number of stochastically labeled and non-labeled features determined in (d) within the region of the array comprises calculating the number of the target molecule based on the number of stochastically labeled features, the number of non-labeled features within the region of the array, and the predictions of the Poisson distribution.

12. The method of claim 1, further comprising calculating a confidence interval for the calculated number of the target molecule.

13. The method of claim 5, wherein automatically locating the one or more of plurality of stochastically labeled and non-labeled features of the array comprises automatically locating the one or more of the plurality of stochastically labeled and non-labeled features of the array using a predefined set of fiducial features on the array.

14. The method of claim 9, wherein calculating the intensity value statistic for signal and background pixels comprises subtracting the mean or median background intensity from the mean or median signal intensity to determine a background corrected signal intensity value for each array feature.