Quantification of nucleic acid molecules using multiplex pcr

Abstract

Described are novel quantification methods and systems that permit, within the context of multiplex PCR, the quantification of all targets within a single reaction tube. The methods employ quantitation algorithms applied to the amplification profiles of internal calibration controls or standards utilizing a plurality of nucleic acid templates that are amplified within the same reaction tube as the nucleic acid target(s) interrogated.

1 . A method of determining an initial concentration (Q 0 ) of a target nucleic acid template in a sample, the method comprising the steps of:
 a) amplifying said target nucleic acid template and a plurality of standard templates, each of said standard templates present at a different, known copy number, in a nucleic acid amplification reaction regimen comprising a plurality of cycles of primer annealing, primer elongation and strand dissociation, wherein said reaction is performed using a reaction mixture comprising said target nucleic acid template and said plurality of standard templates;   b) measuring, at plural cycles of said amplifying regimen, signals from said target template and each of said plurality of standard templates after nucleic acid species in an aliquot taken from the reaction mixture are separated, wherein said measuring generates a set of measurements for each said template;   c) estimating a cycle threshold (C t ) value for each of said plurality of standard templates and said target template, wherein the C t  value is estimated by a method comprising the steps, for each set of measurements for each said template, of:
 i) compiling a candidate list of all sets of consecutive signal measurements with cardinality 3 or greater; 
 ii) removing outliers based on those signal measurements from said candidate list; 
 iii) computing a best-fitting line for an amplification curve for each set of measurements, wherein the best-fitting line correlates a signal measurement to a cycle number; 
 iv) computing a fitness score for each of said target template and said plurality of standard templates, wherein the fitness score is computed as a weighted sum of quality attributes comprising a correlation coefficient determined from the best-fitting line, the cardinality of the candidate data set, and proximity to the signal threshold; 
 v) selecting the set with the highest fitness score for each of said target template and said plurality of standard templates; and 
 vi) computing C t  from the selected set of each of said target template and said plurality of standard templates, whereby C t  is computed by: 
 A) choosing a threshold value for a signal measurement; and 
 B) determining the intercept of the threshold value on the best-fitting line, wherein the C t  is equal to the corresponding cycle number at the intercept. 
   d) generating a standard curve, by plotting C t  values estimated in step (c) on the y axis for each of said plurality of said standard templates versus the log of said known copy number on the x axis for each said standard template; and   e) calculating an initial concentration (Q 0 ) from the standard curve generated in step (d) for said target nucleic acid template.   
     
     
         2 . The method of  claim 1 , wherein the efficiency of amplification is similar for said plurality of standards and said target template. 
     
     
         3 . The method of  claim 1 , wherein the difference in the efficiency of amplification between said plurality of standards and said target template is greater than or equal to 0.1. 
     
     
         4 . The method of  claim 3 , wherein a normalized estimate of the initial concentration (Q 0 ′) for said target nucleic acid template is computed using:
 log(Q 0 ′)=log(Q 0 )*(E R /E T ) Ct , where E T  is the efficiency of said target nucleic acid template and E R  is the efficiency of said plurality of standards. 
 
     
     
         5 . The method of  claim 1 , wherein the outliers comprise the signal measurements from said candidate list for which peak area measurement is less than peak area measurement at the previous cycle. 
     
     
         6 . The method of  claim 1 , wherein said measuring comprises measurement of a fluorescent signal. 
     
     
         7 . The method of  claim 1 , wherein said nucleic acid species are separated by capillary electrophoresis. 
     
     
         8 . The method of  claim 1 , wherein the initial concentration (Q 0 ) for said target template is calculated by solving the equation 
       
         
           
             
               
                 C 
                 
                   t 
                    
                   
                     ( 
                     target 
                     ) 
                   
                 
               
               = 
               
                 
                   [ 
                   
                     I 
                     - 
                     
                       log 
                        
                       
                         ( 
                         
                           Q 
                           0 
                         
                         ) 
                       
                     
                   
                   ] 
                 
                 
                   log 
                    
                   
                     ( 
                     
                       E 
                       R 
                     
                     ) 
                   
                 
               
             
           
         
       
       for Q 0 , wherein I is the X intercept of said standard curve and E R  is the efficiency obtained from said standard curve. 
     
     
         9 . The method of  claim 1 , wherein said measurements of step (b) are entered into a computer-readable physical memory, and wherein steps (c)-(e) are performed by a computer processor executing instructions, encoded on a computer-readable physical memory, for performing such steps. 
     
     
         10 . The method of  claim 1 , wherein each of said standard templates and said target nucleic acid are amplified by the same pair of oligonucleotide primers. 
     
     
         11 . The method of  claim 1 , wherein said measuring step (b) is performed on a plurality of aliquots taken from said reaction mixture at respective plural cycles during said amplifying regimen. 
     
     
         12 . A method of determining initial concentrations (Q 0 ) of a plurality of different target nucleic acid templates in a sample, the method comprising the steps of:
 a) amplifying the plurality of different target nucleic acid templates and a plurality of sets of standard templates, each of said standard templates in a set present at a different, known copy number, in a nucleic acid amplification reaction regimen comprising a plurality of cycles of primer annealing, primer elongation and strand dissociation, wherein said reaction is performed using a reaction mixture comprising said target nucleic acid templates and said plurality of standard templates, wherein the efficiency of amplification of at least one of the target nucleic acid templates is different from the efficiency of amplification of the other different target nucleic acid templates, and wherein the efficiency of amplification of each of the different target nucleic acid templates is similar to the efficiency of amplification of at least one set of the standard templates;   b) measuring, at plural cycles of said amplifying regimen, signals from said target templates and each of said standard templates after nucleic acid species in an aliquot taken from the reaction mixture are separated, wherein said measuring generates a set of measurements for each said template;   c) estimating a cycle threshold (C e ) value for each of said standard templates in each set and said target templates;   d) for each set of said standard templates, generating a standard curve, by plotting C t  values estimated in step (c) on the y axis for each of said standard templates in the set versus the log of said known copy number on the x axis for each said standard template in the set; and   e) for each of said target templates, calculating an initial concentration (Q 0 ) from the standard curve generated in step (d) based on a set of said standard templates having the efficiency of amplification similar to that of the target template.   
     
     
         13 . The method of  claim 12 , wherein the C t  value is estimated by a method comprising the steps, for each set of measurements for each said template, of:
 i) compiling a candidate list of all sets of consecutive signal measurements with cardinality 3 or greater;   ii) removing outliers based on those signal measurements from said candidate list;   iii) computing a best-fitting line for an amplification curve for each set of measurements, wherein the best-fitting line correlates a signal measurement to a cycle number;   iv) computing a fitness score for each of said target templates and said standard templates, wherein the fitness score is computed as a weighted sum of quality attributes comprising a correlation coefficient determined from the best-fitting line, the cardinality of the candidate data set, and proximity to the signal threshold;   v) selecting the set with the highest fitness score for each of said target templates and said standard templates; and   vi) computing C t  from the selected set for each of said target templates and said standard templates, wherein C t  is computed by:   A) choosing a threshold value for a signal measurement; and   B) determining the intercept of the threshold value on the best-fitting line, whereby the C t  is equal to the corresponding cycle number at the intercept.   
     
     
         14 . The method of  claim 13 , wherein the outliers comprise the signal measurements from said candidate list for which peak area measurement is less than peak area measurement at the previous cycle. 
     
     
         15 . The method of  claim 13 , wherein the best-fitting line is computed by linear regression, said line representing a log-linear amplification curve described by the equation (1):
   log(measured value)= C   0   +EC      wherein C is the cycle number, C 0  is the X intercept and E is the slope of the line; and wherein C t  is computed by:   A) choosing a threshold value for log(measured value) of equation (1),   B) solving equation (1) for C when log(measured value) equals the chosen threshold value, and   C) setting C t  equal to the solved value of C.   
     
     
         16 . The method of  claim 12 , wherein said measuring comprises measurement of a fluorescent signal. 
     
     
         17 . The method of  claim 12 , wherein said nucleic acid species are separated by capillary electrophoresis. 
     
     
         18 . The method of  claim 12 , wherein the initial concentration (Q 0 ) for each of said target templates is calculated by solving the equation 
       
         
           
             
               
                 C 
                 
                   t 
                    
                   
                     ( 
                     target 
                     ) 
                   
                 
               
               = 
               
                 
                   [ 
                   
                     I 
                     - 
                     
                       log 
                        
                       
                         ( 
                         
                           Q 
                           0 
                         
                         ) 
                       
                     
                   
                   ] 
                 
                 
                   log 
                    
                   
                     ( 
                     
                       E 
                       R 
                     
                     ) 
                   
                 
               
             
           
         
       
       for Q 0 , wherein I is the X intercept of said standard curve generated in step (d) based on a set of said standard templates having the efficiency of amplification similar to that of the target template and E R  is the efficiency obtained from said standard curve. 
     
     
         19 . The method of  claim 12  wherein said measurements of step (b) are entered into a computer-readable physical memory, and wherein steps (c)-(e) are performed by a computer processor executing instructions, encoded on a computer-readable physical memory, for performing such steps. 
     
     
         20 . A non-transitory computer-readable physical storage medium comprising instructions, that when executed by a processor, cause the processor to perform a procedure comprising the steps of:
 a) receiving a plurality of measurements obtained from a plurality of standard templates;   b) receiving a plurality of measurements obtained from a target template;   c) estimating a cycle threshold (C t ) value for each of said plurality of standard templates and said target template, wherein the C t  value is estimated by a method comprising the steps, for each set of measurements for each said template, of:
 i) compiling a candidate list of all sets of consecutive signal measurements with cardinality 3 or greater; 
 ii) removing outliers based on those signal measurements from said candidate list; 
 iii) computing a best-fitting line for an amplification curve for each set of measurements, wherein the best-fitting line correlates a signal measurement to a cycle number; 
 iv) computing a fitness score for each of said target template and said plurality of standard templates, wherein the fitness score is computed as a weighted sum of quality attributes comprising a correlation coefficient determined from the best-fitting line, the cardinality of the candidate data set, and proximity to the signal threshold; 
 v) selecting the set with the highest fitness score for each of said target template and said plurality of standard templates; and 
 vi) computing C t  from the selected set of each of said target template and said plurality of standard templates, whereby C t  is computed by:
 A) choosing a threshold value for a signal measurement; and 
 B) determining the intercept of the threshold value on the best-fitting line, wherein the C t  is equal to the corresponding cycle number at the intercept. 
 
   d) generating a standard curve by plotting C t  values estimated in step (c) on the y axis for each of said plurality of said standard templates versus the log of said known copy number on the x axis for each said standard template; and   e) calculating an initial concentration (Q 0 ) from the standard curve generated from step (d) for said target nucleic acid template.   
     
     
         21 . A system comprising a computer processor, and instructions that cause the processor to perform a procedure comprising the steps of:
 a) receiving a plurality of measurements obtained from a plurality of standard templates;   b) receiving a plurality of measurements obtained from a target template;   c) estimating a cycle threshold (C t ) value for each of said plurality of standard templates and said target template, wherein the C t  value is estimated by a method comprising the steps, for each set of measurements for each said template, of:
 i) compiling a candidate list of all sets of consecutive signal measurements with cardinality 3 or greater; 
 ii) removing outliers based on those signal measurements from said candidate list; 
 iii) computing a best-fitting line for an amplification curve for each set of measurements, wherein the best-fitting line correlates a signal measurement to a cycle number; 
 iv) computing a fitness score for each of said target template and said plurality of standard templates, wherein the fitness score is computed as a weighted sum of quality attributes comprising a correlation coefficient determined from the best-fitting line, the cardinality of the candidate data set, and proximity to the signal threshold; 
 v) selecting the set with the highest fitness score for each of said target template and said plurality of standard templates; and 
 vi) computing C t  from the selected set of each of said target template and said plurality of standard templates, whereby C t  is computed by:
 A) choosing a threshold value for a signal measurement; and 
 B) determining the intercept of the threshold value on the best-fitting line, wherein the C t  is equal to the corresponding cycle number at the intercept. 
 
   d) generating a standard curve by plotting C t  values estimated in step (c) on the y axis for each of said plurality of said standard templates versus the log of said known copy number on the x axis for each said standard template; and   e) calculating an initial concentration (Q 0 ) from the standard curve generated in step (d) for said target nucleic acid template.