Methods for accurate component intensity extraction from separations-mass spectrometry data

• US 20050255606A1
• Filed: 05/13/2004
• Published: 11/17/2005
• Est. Priority Date: 05/13/2004
• Status: Abandoned Application

• Current Assignee: Vern Norviel, V&M TCA, LP
• Original Assignee: Biospect Incorporated (Response Genetics, Inc)
• Inventors: Ahmed, Zulfikar, Quarato, Jim, Balasingham, Arjuna, De Valpine, David Perry, Brown, Michael, Donoho, David, Bitter, Hans, Heller, Jonathan C.
• Application Number:
  US10/846,996
• Publication Number:
  US 20050255606A1
• Time in Patent Office: Days
• US Class Current: 436/173
• CPC Class Codes: G06F 2218/02 Preprocessing
• CPC Class Codes: H01J 49/0036 Step by step routines describing the handling of the data generated during a measurement
• CPC Class Codes: H01J 49/02 Details
• CPC Class Codes: Y10T 436/24 Nuclear magnetic resonance, electron spin resonance or other spin effects or mass spectrometry

Abstract

The present invention discloses methods for deconvolving and converting 1D mass spectra to 2D mass spectrum in order to obtain migration time centers and total intensities of the neutral mass envelopes of 2D spectra. The present invention also discloses devices that include a preparation/separation unit coupled to a mass spectrometer unit, and a computer unit capable of deconvolving mass spectra and calculating neutral mass envelopes.

Claims

1. A method comprising:

generating a mass spectrum;

providing a lineshape for said mass spectrum; and

deconvolving said mass spectrum with said lineshape.

2. The method of claim 1 wherein said mass spectrum is a 2D separation-mass spectrum.

3. The method of claim 1 wherein said mass spectrum is a mass-to-charge ratio mass spectrum.

4. The method of claim 1 wherein said generating step involves the use of a mass spectrometer.

5. The method of claim 4 wherein said mass spectrometer is a time-of-flight mass spectrometer or a fourier transform ion cyclotron mass spectrometer.

6. The method of claim 5 wherein said mass spectrometer is a time-of-flight mass spectrometer.

7. The method of claim 4 wherein the mass spectrometer collects tandem mass spectrometry data.

8. The method of claim 4 wherein said mass spectrometer comprises an ion source selected from the group consisting of:

an ESI, a nano-ESI, atmospheric pressure chemical ionization, matrix-assisted laser desorption ionization, surface-enhanced laser desorption ionization, desorption ionization on silicon, fast atom/ion bombardment, electron ionization, and chemical ionization.

9. The method of claim 8 wherein said ion source is an ESI.

10. The method of claim 4 wherein said mass spectrometer is coupled to a separation device.

11. The method of claim 1 further comprising a step of separating a sample prior to generating said mass spectrum.

12. The method of claim 11 wherein said separating is preformed by electrophoresis or high performance liquid chromatography.

13. The method of claim 11 wherein said separating is performed by microfluidic chip.

14. The method of claim 11 wherein said separating device separates a composition having a molecular weight selected from the group consisting of less than 2 kDa, less than 30 kDa, less than 50 kDa, 50 Da-150 kDa, and more than 150 kDA.

15. The method of claim 11 wherein said lineshape is determined based upon at least one physical parameter of the separating step or the generating mass spectrum step.

16. The method of claim 1 wherein said lineshape is determined from raw data.

17. The method of claim 1 further comprising the step of estimating one or more parameters that determine said lineshape.

18. The method of claim 6 further comprising the step of scaling said lineshape as a function of time-of-flight.

19. The method of claim 6 wherein said lineshape varies deterministically along a time-of-flight axis.

20. The method of claim 19 wherein the width of the lineshape varies linearly or quadratically as a function of time-of-flight.

21. The method of claim 20 wherein the linear or quadratic parameters are calculated from data using a parametric model of lineshape.

22. The method of claim 21 wherein said parametric model of lineshape is determined using a model of said lineshape that comprises initial position and energy distribution of ions.

23. The method of claim 21 wherein said parametric model of lineshape is gaussian.

24. The method of claim 21 wherein said parametric model of lineshape is student-t distribution.

25. The method of claim 21 wherein said parametric model of lineshape is determined by computer simulation of said mass spectrometer.

26. The method of claim 1 wherein said deconvolving step comprises an algorithm selected from the group consisting of basis pursuit (one-norm penalty), Tikhonov regularization (two-norm penalty), maximum entropy (entropy penalty), and parametric deconvolution.

27. The method of claim 1 wherein said deconvolving step involves the use of basis pursuit algorithm.

28. The method of claim 1 wherein said deconvolving step further comprises estimating noise level.

29. The method of claim 28 wherein said noise level is used in an objective function calculation for said deconvolving step.

30. The method of claim 1 wherein said deconvolving step further comprises use of fast wavelet transform for convolution calculation.

31. The method of claim 1 wherein the said deconvolving step yields data with increased resolution.

32. The method of claim 31 wherein said resolution is increased by at least 1.5.

33. The method of claim 1 wherein the said deconvolving step reduces noise.

34. The method of claim 33 wherein said deconvolving step reduces noise by modifying the objective function to be a penalized log-likelihood function rather than a penalized least-squares problem.

35. The method of claim 1 wherein the said deconvolving step increases signal-to-noise ratio.

36. The method of claim 35 wherein said signal-to-noise ratio is increased by at least 2, 5, 10,or 50.

37. The method of claim 1 further comprising the step of correcting deconvolved spectrum using isotope distribution data to group deconvolved peaks into isotopic clusters.

38. The method of claim 37 wherein said isotope data is modeled as a binomial distribution with parameters N and p, where N is the approximate number of carbons and p is the probability of occurrence of carbon-13 isotope.

39. The method of claim 38 wherein the approximate number of carbons is estimated by regression of number of carbons from a set of known peptides.

40. The method of claim 38 wherein the probability of occurrence of carbon-13 in proteins and peptides in sample is estimated from data.

41. The method of claim 37 wherein said isotope distribution data and the lineshape are used to calculate a charge state of an envelope.

42. The method of claim 1 further comprising the step of descaling deconvolving mass spectrum.

43. The method of claim 1 further comprising the step of converting 1D mass spectrum to 2D mass spectrum.

44. The method of claim 43 further comprising the step of conducting 2D cluster analysis to determine centroid location for each envelope.

45. The method of claim 1 further comprising the step of calculating a charge state for an envelope.

46. The method of claim 45 wherein the charge state is calculated using the width of the lineshape and the width of an unresolved enveloped peak in the raw spectrum.

47. The method of claim 45 wherein the charge state is calculated using the width of the lineshape and the width of a deconvolved envelope with the lineshape.

48. The method of claim 45 wherein the charge state is calculated using the spacing between peaks in a corrected deconvolved output within a cluster.

49. The method of claim 1 further comprising the step of creating a list of 2D peaks in the spectrum by their positions and total intensities.

50. The method of claim 1 further comprising the step of creating a list of neutral mass components by their migration times and total intensities.

51. The method of claim 1 further comprising the step of aligning a plurality of lists of neutral masses from multiple 2D mass spectra or a plurality of lists of 2D peaks from multiple 2D mass spectra, wherein said lists provide location and total intensity for each neutral mass or 2D peak.

52. The method of claim 51 wherein the list of 2D peaks is collapsed to a neutral mass component list.

53. A method comprising:

creating a list of 2D peaks derived from a deconvolved mass spectrum; and

aligning a plurality of such lists.

54. A method comprising

creating a list of neutral mass components derived from a mass spectrum;

and

aligning a plurality of such lists.

55. A method for diagnosing a mammal comprising the steps of:

obtaining a sample from said mammal;

analyzing the sample with a device that performs separation and mass spectrometry;

determining a list of 2D peaks derived from said separation and mass spectrometry; and

identifying the existence or lack of existence of a 2D peak or a pattern of 2D peaks.

56. A method for diagnosing a mammal comprising the steps of:

obtaining a sample from said mammal;

analyzing the sample with a device that performs separation and mass spectrometry;

determining a list of neutral mass components in said sample; and

identifying the existence or lack of existence of a neutral mass component or a pattern.

57. A method for diagnosing a disease state in a mammal comprising the steps of:

obtaining a sample from said mammal;

performing separations on the said sample;

generating a mass spectrum from said sample;

providing a lineshape for said spectrum; and

deconvolving said spectrum with said lineshape.

58. The method of claim 57 wherein said mass spectrum is a 2D mass spectrum.

59. The method of claim 57 wherein said sample is a liquid sample selected from the group consisting of urine, nasal discharge, vaginal discharge, mucus, lymph, blood, serum, plasma, saliva, and tears.

60. The method of claim 57 wherein said generating step involves the use of a mass spectrometer.

61. The method of claim 60 wherein said mass spectrometer is selected from the group consisting of a time-of-flight mass spectrometer, a time-of-flight reflectron mass spectrometer, a Quad time-of-flight mass spectrometer, and a Fourier transform ion cyclotron mass spectrometer.

62. The method of claim 60 wherein said mass spectrometer is a time-of-flight mass spectrometer.

63. The method of claim 60 wherein the mass spectrometer collects tandem mass spectrometry data.

64. The method of claim 60 wherein said mass spectrometer comprises an ion source selected from the group consisting of:

an ESI, a nano-ESI, atmospheric pressure chemical ionization, matrix-assisted laser desorption ionization, surface-enhanced laser desorption ionization, desorption ionization on silicon, fast atom/ion bombardment, electron ionization, and chemical ionization.

65. The method of claim 64 wherein said ion source is an ESI.

66. The method of claim 62 wherein said mass spectrometer is coupled to a separation device.

67. The method of claim 66 wherein said separation device performs electrophoresis or high performance liquid chromatography.

68. The method of claim 67 wherein said separation device performs electrophoresis.

69. The method of claim 66 wherein said separation device is a microfluidic chip.

70. The method of claim 59 wherein said lineshape is determined based on at least one physical parameter of a separation-mass spectrometer device.

71. The method of claim 59 wherein said lineshape is determined from raw data.

72. The method of claim 59 further comprising the step of estimating one or more parameters that determine said lineshape.

73. The method of claim 59 further comprising the step of scaling said lineshape along a time-of-flight axis.

74. The method of claim 59 wherein said lineshape varies deterministically along a time-of-flight axis.

75. The method of claim 59 wherein width of the lineshape varies according to a linear parameter or a quadratic parameter as a function of time-of-flight.

76. The method of claim 75 wherein the linear or quadratic parameter is calculated from data using a parametric model of lineshape.

77. The method of claim 76 wherein said parametric model of lineshape is determined using a model of said lineshape that comprises initial position and energy distribution of ions.

78. The method of claim 76 wherein said parametric model of lineshape is gaussian.

79. The method of claim 76 wherein said parametric model of lineshape is Student-t distribution.

80. The method of claim 76 wherein said parametric model of lineshape is determined by computer simulation of said mass spectrometer.

81. The method of claim 59 wherein said deconvolving step comprises using an algorithm selected from the group consisting of basis pursuit (one-norm penalty), Tikhonov regularization (two-norm penalty), maximum entropy (entropy penalty), and parametric deconvolution.

82. The method of claim 59 wherein said deconvolving step comprises using basis pursuit algorithm.

83. The method of claim 59 wherein said deconvolving step further comprises estimating noise level.

84. The method of claim 83 wherein said noise level is used in an objective function calculation for said deconvolving step.

85. The method of claim 59 wherein said deconvolving step further comprises of the use of fast wavelet transform for convolution calculation.

86. The method of claim 59 wherein the said deconvolving step yields data with increased resolution.

87. The method of claim 86 wherein said resolution is increased by at least 1.5.

88. The method of claim 59 wherein the said deconvolving step reduces noise.

89. The method of claim 88 wherein said deconvolving step reduces noise by modifying the objective function to be a penalized log-likelihood function rather than a penalized least-squares problem.

90. The method of claim 59 wherein the said deconvolving step increases signal-to-noise ratio.

91. The method of claim 90 wherein said signal-to-noise ratio is increased by at least 2, 5, 10, or 50.

92. The method of claim 59 wherein deconvolved spectrum is corrected by using isotope distribution data to group deconvolved peaks into isotopic clusters.

93. The method of claim 92 wherein said isotope data is modeled as a binomial distribution with parameters N and p, where N is the approximate number of carbons and p is the probability of occurrence of carbon-13 isotope.

94. The method of claim 93 wherein the approximate number of carbons is estimated by regression of number of carbons from a set of known peptides.

95. The method of claim 93 wherein the probability of occurrence of carbon-13 is estimated from the spectrum.

96. The method of claim 92 wherein said isotope distribution data and the lineshape are used to calculate a charge state of an envelope.

97. The method of claim 59 wherein deconvolved spectrum resulting from said deconvolving step is descaled.

98. The method of claim 59 further comprising the step of descaling output from said deconvolving step.

99. The method of claim 92 wherein corrected deconvolved spectra are submitted to a 2D cluster analysis to determine centroid location for each envelope.

100. The method of claim 59 further comprising the step of conducting 2D cluster analysis to determine centroid location for each envelope.

101. The method of claim 59 further comprising the step of calculating for each peak one or more data points selected from the group consisting of:

mass-to-charge, mass, monoisotopic abundance, total abundance, migration time centroid, charge state, and migration time width.

102. The method of claim 101 wherein the charge state is calculated using the width of the lineshape and the width of the unresolved enveloped peak in the raw spectrum.

103. The method of claim 101 wherein the charge state is calculated using the width of the lineshape and the width of the deconvolved envelope with the lineshape.

104. The method of claim 101 wherein the charge state is calculated using the spacing between the peaks in a corrected deconvolved output within a cluster.

105. The method of claim 59 further comprising the step of creating a list of 2D peaks in the spectrum by their positions and total intensities.

106. The method of claim 59 further comprising the step of creating a list of neutral mass components by their migration times and total intensities.

107. The method of claim 59 further comprising the step of aligning a plurality of lists of neutral masses from multiple 2D mass spectra or a plurality of lists of 2D peaks from multiple 2D mass spectra, wherein said lists provide location and total intensity for each neutral mass or 2D peak.

108. The method of claim 107 wherein the list of 2D peaks is collapsed to a neutral mass component list.

109. The method of claim 108 wherein the presence or absence of a neutral mass or a 2D peak is indicative of a disease state.

110. The method of claim 108 wherein the presence or absence of a pattern of neutral mass or 2D peaks is indicative of disease state.

111. The method of claim 59 wherein the disease state is selected from the group consisting of a neoplastic disease, an immunologic disease, an endocrine disease, a metabolic disease, or a cardiovascular disease.

112. The method of claim 111 wherein the disease state is a neoplastic disease.

113. The method of claim 112 wherein the neoplastic disease is selected from the group consisting of:

brain cancer, breast cancer, bone cancer, cancer of the larynx, gallbladder, pancreas, rectum, parathyroid, thyroid, adrenal, neural tissue, head and neck, colon, stomach, bronchi, kidneys, basal cell carcinoma, squamous cell carcinoma of both ulcerating and papillary type, metastatic skin carcinoma, osteo sarcoma, Ewing'"'"'s sarcoma, veticulum cell sarcoma, myeloma, giant cell tumor, small-cell lung tumor, gallstones, islet cell tumor, primary brain tumor, acute and chronic lymphocytic and granulocytic tumors, hairy-cell tumor, adenoma, hyperplasia, medullary carcinoma, pheochromocytoma, mucosal neuronms, intestinal ganglloneuromas, hyperplastic corneal nerve tumor, marfanoid habitus tumor, Wilm'"'"'s tumor, seminoma, ovarian tumor, leiomyomater tumor, cervical dysplasia and in situ carcinoma, neuroblastoma, retinoblastoma, soft tissue sarcoma, malignant carcinoid, topical skin lesion, mycosis fungoide, rhabdomyosarcoma, Kaposi'"'"'s sarcoma, osteogenic and other sarcoma, malignant hypercalcemia, renal cell tumor, polycythermia vera, adenocarcinoma, glioblastoma multiforma, leukemias, lymphomas, malignant melanomas, skin cancer, leukemia, prostate cancer, liver cancer, lung cancer, and epidermoid carcinomas.

114. The method of claim 59 wherein said mammal is a human.

115. The method of claim 59 further comprising a step of adding acid to the sample that will denature a protein in the sample.

116. The method of claim 59 further comprising a step of filtering the sample.

117. The method of claim 116 wherein said filtering step eliminates from analysis compositions greater than 30 kDa.

118. The method of claim 117 further comprising a step of concentrating said filtrate with a reverse phase column.

119. An apparatus comprising

a separation unit;

a mass spectrometer; and

a computer system, wherein said computer system can perform the functions of;

providing a lineshape for a mass spectrum; and

deconvolving said mass spectrum with said lineshape.

120. The apparatus of claim 119 wherein said separation unit performs electrophoresis or high performance liquid chromatography.

121. The apparatus of claim 119 wherein said separation unit is a microfluidic chip.

122. The apparatus of claim 121 wherein said mass spectrometer is selected from the group consisting of a time-of-flight mass spectrometer, a time-of-flight reflectron mass spectrometer, a Quad time-of-flight mass spectrometer, and a Fourier transform ion cyclotron mass spectrometer.

123. The method of claim 119 wherein said mass spectrometer is a time-of-flight mass spectrometer.

124. The method of claim 119 wherein the mass spectrometer collects tandem mass spectrometry data.

125. The method of claim 123 wherein said mass spectrometer comprises an ion source selected from the group consisting of:

an ESI, a nano-ESI, atmospheric pressure chemical ionization, matrix-assisted laser desorption ionization, desorption ionization on silicon, fast atom/ion bombardment, electron ionization, and chemical ionization.

126. The method of claim 125 wherein said ion source is an ESI or a nano-ESI.

127. The method of claim 123 wherein said separation unit and said mass spectrometer are connected online.

128. The method of claim 123 wherein said computer system further performs the steps of scaling and descaling a mass spectra.

129. The method of claim 123 wherein said lineshape is determined based on at least one physical parameter of the separation unit or the mass spectrometer.

130. The method of claim 123 wherein said lineshape is determined based on raw data.

131. The method of claim 123 further comprising the step of estimating one or more parameters that determine said lineshape.

132. The method of claim 127 wherein said computer unit further performs the function of scaling said lineshape along a time-of-flight axis.

133. The method of claim 127 wherein said lineshape varies deterministically along a time-of-flight axis.

134. The method of claim 119 wherein the width of the lineshape varies according to a linear or a quadratic parametric as a function of time-of-flight.

135. The method of claim 134 wherein the linear or quadratic parameter is calculated from data using a parametric model of lineshape.

136. The method of claim 135 wherein said parametric model of lineshape is determined using a model of said lineshape that comprises initial position and energy distribution of ions.

137. The method of claim 135 wherein said parametric model of lineshape is gaussian.

138. The method of claim 135 wherein said parametric model of lineshape is Student-t distribution.

139. The method of claim 135 wherein said parametric model of lineshape is determined by computer simulation of said mass spectrometer.

140. The method of claim 119 wherein said computer deconvolves using an algorithm selected from the group consisting of basis pursuit (one-norm penalty), Tikhonov regularization (two-norm penalty), maximum entropy (entropy penalty), and parametric deconvolution.

141. The method of claim 119 wherein said computer deconvolving using basis pursuit algorithm.

142. The method of claim 119 wherein said deconvolving further comprises estimating noise level.

143. The method of claim 142 wherein said noise level is used in an objective function calculation for said deconvolving step.

144. The method of claim 119 wherein said deconvolving further comprises of the use of fast wavelet transform for convolution calculation.

145. The method of claim 119 wherein the said deconvolution step yields data with increased resolution.

146. The method of claim 145 wherein said resolution is increased by at least 1.5.

147. The method of claim 119 wherein the said deconvolving step reduces noise.

148. The method of claim 147 wherein the post-deconvolution mass spectrum has peak intensities that deviate from true area under the raw noiseless peak by at most 30%.

149. The method of claim 119 wherein the said deconvolution algorithm increases signal-to-noise ratio.

150. The method of claim 149 wherein said signal-to-noise ratio is increased by at least 2, 5, 10, or 50.

151. The method of claim 119 wherein said computer unit further corrects deconvolved spectrum using isotope distribution data.

152. The method of claim 151 wherein said isotope data is modeled as a binomial distribution with parameters N and p, where N is the approximate number of carbons and p is the probability of occurrence of carbon-13 isotope.

153. The method of claim 152 wherein the approximate number of carbons is estimated by regression of number of carbons from a set of known peptides.

154. The method of claim 152 wherein the probability of occurrence of carbon-13 is estimated from the spectrum.

155. The method of claim 151 wherein said isotope distribution data and the lineshape are used to calculate a charge state of an envelope.

156. The method of claim 119 wherein said computer system further performs the function of descaling output from said deconvolving step.

157. The method of claim 151 wherein said computer system further performs 2D cluster analysis on said corrected deconvolved spectra to determine centroid location for each envelope.

158. The method of claim 119 wherein said computer system further performs the step of calculating for each peak its mass-to-charge, mass, monoisotopic abundance, total abundance, migration time centroid, charge state, or migration time width.

159. The method of claim 158 wherein the charge state is calculated using the width of the lineshape and the width of the unresolved enveloped peak in the raw spectrum.

160. The method of claim 158 wherein the charge state is calculated using the width of the lineshape and the width of the deconvolved envelope with the lineshape.

161. The method of claim 158 wherein the charge state is calculated using the spacing between the peaks in a corrected deconvolved output within a cluster.

162. The method of claim 119 herein said computer system creates a list of 2D peaks in the spectrum by their positions and total intensities.

163. The method of claim 119 wherein said computer system creates a list of neutral mass components by their migration times and total intensities.

164. The method of claim 119 wherein said computer system aligns a plurality of lists of neutral masses or a plurality of lists of 2D peaks, wherein said lists provide location and total intensity for each neutral mass or 2D peak.

165. The method of claim 164 wherein said computer system can further collapse the list of 2D peaks to a list of neutral mass components.

166. A method comprising the steps of:

acidifying a sample;

providing a sample;

separating a composition from said sample;

analyzing separated sample using a mass analyzer;

scaling a mass spectrum generated by the mass analyzer;

deconvolving the scaled mass spectrum;

descaling the deconvolved mass spectrum;

deisotoping the descaled mass spectrum;

decharging the de-isotoped mass spectrum;

providing a list of 2D peaks;

providing a list of neutral mass components;

aligning a plurality of 2D peak lists or a plurality of neutral mass component lists.

167. The method of claim 166 further providing the step of acidifying said sample.

168. The method of claim 166 wherein said sample is derived from a mammal.

169. The method of claim 166 wherein said separating step separates samples <

30 kDa.

170. The method of claim 166 wherein said deconvolving step involves the use of an algorithm selected from the group consisting of basis pursuit (one-norm penalty), Tikhonov regularization (two-norm penalty), maximum entropy (entropy penalty), and parametric deconvolution.

171. The method of claim 166 further comprising the step of providing a lineshape to said mass spectrum.

172. The method of claim 171 wherein said lineshape is determined from raw data.

173. The method of claim 171 further comprising the step of estimating one or more parameters that determine said lineshape.

174. The method of claim 171 wherein width of said lineshape varies linearly or quadratically as a function of time-of-flight.

175. The method of claim 171 wherein said lineshape varies deterministically along a time-of-flight axis.

176. The method of claim 171 wherein said scaling step scales said spectrum and lineshape along a time-of-flight axis.

177. The method of claim 172 wherein said deconvolving step further comprises estimating noise level.

178. The method of claim 177 wherein said noise level is used in an objective function calculation for said deconvolving step.

179. The method of claim 172 wherein said deconvolving step further comprises of the use of fast wavelet transform for convolution calculation.

180. The method of claim 172 wherein the said deconvolving step yields data with increased resolution.

181. The method of claim 180 wherein said resolution is increased by at least 1.5.

182. The method of claim 172 wherein the said deconvolving step reduces noise.

183. The method of claim 182 wherein said deconvolving step reduces noise by modifying the objective function to be a penalized log-likelihood function rather than a penalized least-squares problem.

184. The method of claim 172 wherein the said deconvolving step increases signal-to-noise ratio.

185. The method of claim 184 wherein said signal-to-noise ratio is increased by at least 2, 5, 10, or 50.

186. The method of claim 172 wherein deconvolved spectrum is corrected by using isotope distribution data to group deconvolved peaks into isotopic clusters.