Method for reducing cross-talk within DNA data

• US 6,598,013 B1
• Filed: 07/31/2001
• Issued: 07/22/2003
• Est. Priority Date: 07/31/2001
• Status: Expired due to Fees

• Current Assignee: University of Toledo
• Original Assignee: The University Of Maine
• Inventors: Domnisoru, Cristian, Musavi, Mohamad
• Primary Examiner(s): WACHSMAN, HAL D
• Application Number:
  US09/918,885
• Time in Patent Office: 721 Days
• Field of Search: 702/191, 702/19, 702/20, 702/22, 702/23, 702/25, 702/27, 702/28, 702/30-32, 702/66, 702/70, 702/71, 702/73-77, 702/106, 702/124, 702/126, 702/127, 702/183, 702/189, 702/190, 702/193, 702/195-197, 702/179, 702/FOR.170, 702/FOR.171, 702/FOR.103, 702/FOR.104, 702/FOR.105, 702/FOR.108, 702/FOR.110, 702/FOR-115, 702/FOR.121, 702/FOR.131, 702/FOR.134, 702/FOR.164, 702/FOR.166, 702/FOR.168, 706/924, 204/430, 204/451, 204/456, 204/457, 204/406, 204/407, 204/461, 204/606, 204/612, 382/129, 435/6, 435/91.1, 435/91.2, 435/287.1, 536/24.3, 436/172, 436/63, 422/82.07, 422/82.08, 422/68.1, 422/62, 356/344
• US Class Current: 702/191
• CPC Class Codes: G01N 27/44717 Arrangements for investigating the separated zones, e.g. localising zones
• CPC Class Codes: G01N 27/44721 by optical means

Abstract

Raw DNA data is filtered with a multi-component analysis that is applied to the difference of the signal intensity on each of the raw DNA data signals to remove cross talk between the signals. The analysis is done before any baseline adjustment of the raw DNA data. Instead, the baseline adjustment occurs after the raw DNA data has been filtered. Additionally, an additional processing step is applied to the data to account for the non-linear nature of cross talk filtering. The additional processing step involves combining the signal with its derivative to account for the correlation of each of the data signals with the other three data signals.

Claims

1. A method for enhancing DeoxyriboNucleic Acid (DNA) raw data comprising the steps of:

(a) providing an apparatus for collecting DNA data from dye-labeled DNA fragments, the DNA data being divided between a plurality of channels;

(b) passing the DNA data contained in each of the plurality of channels through a filter to reduce cross-talk between DNA data contained in each of the channels; and

(c) adjusting the baseline of the DNA data contained in each of the channels.

2. The method according to claim 1 wherein the reduction of cross talk between each of the channels in step (b) includes the steps of:

(b1) determining difference values for the signals in each channel that correspond to the DNA data by subtracting the magnitudes of the signals in each of the channels at two consecutive sampling instants;

(b2) applying a multi-component analysis to the difference values obtained in step (b1) to deconvolute the DNA data contained in the signals; and

(b3) recombining the deconvoluted difference data with the corresponding signals at the specific sampling instant to obtain the signal intensity.

3. The method according to claim 2 wherein the multi-component analysis in step (b2) includes multiplying the signals corresponding to the DNA data by a constant coefficient transformation matrix M.

4. The method according to claim 3 wherein the multi-component analysis in step (b2) includes the following operation:

${\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_j=\underset{\underset{i=1}{j=1}}{\stackrel{4}{\sum <br><br>}}<br><br>m_{i,j}·<br><br>\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s_j,;<br><br>$

where Δ

s_j represents the variation of the measured signal s_j in each channel that corresponds to the DNA data between two consecutive signal measurements and Δ

s_j represents filtered signal variation with crosstalk removed; and

m_i,j is a constant coefficient indicating the cross talk between measured signal varation s_j and the filtered signal variation Δ

s_j.

5. The method according to claim 4 wherein, prior to step (b), the signal corresponding to the DNA data is passed through a low pass filter.

6. The method according to claim 5 further including, before adjusting the baseline of the DNA data in step (c), passing the signal corresponding to the DNA data in each of the channels through a high pass filter.

7. The method according to claim 6 further including, subsequent to the baseline adjustment in step (c), locating peak values in each channel and reading a DNA sequence from a combination of the DNA data contained in each of the channels.

8. A method for enhancing DeoxyriboNucleic Acid (DNA) raw data comprising the steps of:

(a) providing an apparatus for collecting DNA data from dye-labeled DNA fragments, said data divided between a plurality of channels;

(b) passing the DNA data in each channel through a first filter to reduce cross-talk between DNA data contained in each of the channels;

(c) passing the filtered DNA data in each channel from step (b) through a second filter to reduce any non-linearity remaining after the first filtering process in step (b);

(d) recombining the filtered DNA data in each channel from step (c) with corresponding signals at a specific sampling instant to obtain a filtered signal intensity for each channel; and

(e) adjusting the baseline of the DNA data contained in each of the channels.

9. The method according to claim 8 wherein the second filtering process in step (c) includes the steps of:

(c1) determining derivative values for the filtered DNA data obtained from the first filter in step (b); and

(c2) applying a multi-component analysis to the derivative values obtained in step (c1) to remove non-linear effects remaining after the first filtering process.

10. The method according to claim 9 wherein the multi-component analysis in step (c2) includes multiplying the derivative values obtained in step (c1) by a constant coefficient transformation matrix T.

11. The method according to claim 10 wherein the multi-component analysis in step (c2) includes the following operation:

${\underset{\underset{\_}{\_}}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_i={\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_i+\underset{j\ne <br><br>i}{\underset{j=1}{\stackrel{4}{\sum <br><br>}}}<br><br>t_{i,j}·<br><br>{\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_j^{\mathrm{\prime <br><br>}};<br><br>$

where Δ

s_i represents the variation of the DNA data after the second filtering process in step (c);

Δ

s_i represents the variation of the DNA data after the first filtering process in step (b);

Δ

s_j′

represents the time derivative of Δ

s_i; and

t_i,j is a constant coefficient indicating an approximated linear relationship between intensity Δ

s_i and Δ

s_j′

.

12. The method according to claim 11 wherein the reduction of cross talk between each of the channels in step (b) includes the steps of:

(b1) determining difference values for the DNA data in each channel by subtracting the magnitudes of the DNA data in each channel at two consecutive sampling instants; and

(b2) applying a multi-component analysis to the difference values obtained in step (b1) to deconvolute the DNA data contained in each channel.

13. The method according to claim 12 wherein the multi-component analysis in step (b2) includes multiplying the DNA data by a constant coefficient transformation matrix M.

14. The method according to claim 13 wherein the multi-component analysis in step (b2) includes the following operation:

${\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_j=\underset{\underset{i=1}{j=1}}{\stackrel{4}{\sum <br><br>}}<br><br>m_{i,j}·<br><br>\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s_j,;<br><br>$

where Δ

s_j represents the variation of the measured DNA data s_j between two consecutive DNA data measurements and Δ

s_j represents filtered DNA data variation with crosstalk removed; and

m_i,j is a constant coefficient indicating the cross talk between measured DNA data varation s_j and the filtered signal variation Δ

s_j.

15. The method according to claim 14 wherein, prior to step (b), the DNA data is passed through a low pass filter.

16. The method according to claim 15 further including, before adjusting the baseline of the DNA data in step (e), passing the DNA data through a high pass filter.

17. The method according to claim 16 further including, subsequent to the baseline adjustment in step (e), locating peak values in the DNA data in each channel and reading a DNA sequence from a combination of the DNA data contained in each of the channels.

18. An algorithm for processing DeoxyriboNucleic Acid (DNA) data, the DNA data being divided between a plurality of channels, the algorithm comprising the steps of:

(a) measuring a signal in each channel that corresponds to the DNA data associated with the channel;

(b) determining difference values for the measured signals in each channel by subtracting the magnitudes of the measured signals in each channel at two consecutive sampling instants;

(c) passing the difference values for the measured signals in each of the channels through a first filter to reduce cross-talk between the measured signals contained in the channels that includes the following operation;

${\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_j=\underset{\underset{i=1}{j=1}}{\stackrel{4}{\sum <br><br>}}<br><br>m_{i,j}·<br><br>\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s_j,;<br><br>$

where Δ

s_j represents the variation of the measured signal s_j between two consecutive signal measurements and Δ

s_j represents filtered measured signal variation with crosstalk removed; and

m_i,j is a constant coefficient indicating the cross talk between measured signal varation s_j and the filtered signal variation Δ

s_j;

(d) passing the filtered measured signals from step (c) in each of the channels through a second filter to reduce any non-linearity remaining after the first filtering process in step (c) that includes the following operation;

${\underset{\underset{\_}{\_}}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_i={\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_i+\underset{j\ne <br><br>i}{\underset{j=1}{\stackrel{4}{\sum <br><br>}}}<br><br>t_{i,j}·<br><br>{\underset{\_}{\mathrm{\Delta <br><br>}<br><br>\text{\hspace{1em}<br><br>}<br><br>s}}_j^{\mathrm{\prime <br><br>}};<br><br>$

where Δ

s_i represents the variation of the data signal after the second filtering process;

Δ

s_i represents the variation of the measured signal after the first filtering process in step (c);

Δ

s_j′

represents the time derivative of Δ

s_i; and

t_i,j is a constant coefficient indicating the non-linear relationship between Δ

s_i and Δ

s_j′

;

(e) reconstructing the measured signals in each of the channels; and

(f) adjusting the baseline of the measured signal contained in each of the channels.

19. An algorithm for processing DeoxyriboNucleic Acid (DNA) data, the DNA data being divided between a plurality of channels, the algorithm comprising the steps of:

(a) passing the DNA data in each channel through a first filter to reduce cross-talk between DNA data contained in each of the channels that includes the following operation;

$s_j=\underset{i=1}{\underset{j=1}{\stackrel{4}{\sum <br><br>}}}<br><br>m_{i,j}·<br><br>f_j,;<br><br>$

where s_j represents a signal corresponding to the filtered DNA data and f_j represents fluorescence intensity; and

m_i,j is a constant coefficient indicating the cross talk between intensity signals i and j;

(b) passing the filtered DNA data signal in each of the channels through a second filter to reduce any non-linearity introduced by the first filtering process in step

(a) that includes the following operation;

${\underset{\underset{\_}{\_}}{s}}_i={\underset{\_}{s}}_i+\underset{j\ne <br><br>i}{\underset{j=1}{\stackrel{4}{\sum <br><br>}}}<br><br>t_{i,j}·<br><br>{\underset{\_}{s}}_j^{\mathrm{\prime <br><br>}};<br><br>$

where s_i represents the variation of the DNA data signal after the second filtering process in step (b);

s_i represents the measured signal after the first filtering process in step (a);

s_j′

represents the time derivative of s_i; and

t_i,j is a constant coefficient indicating the non-linear relationship between intensity Δ

s_i and Δ

s_j′

;

(c) reconstructing the DNA data signals in each of the channels; and

(d) adjusting the baseline of the DNA data contained in each of the channels.