ercid_doublecol.py 6.93 KB
Newer Older
Yan's avatar
Yan committed
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
#!/usr/bin/env python3
from rawprasslib import load_raw
from rawautoparams import load_params
from matplotlib import pyplot as plt
import errno
import numpy as np
import os
import time
import scipy.optimize as spopt
import matplotlib.pyplot as plt

# SCRIPT DOES NOT CHECK FOR SANITY OF THE OUTCOME !!!
# THAT'S YOUR JOB !!!

# AE == appearance energy from LCQ obtained by sigmoid fitting
# E_tcid == threshold collision induced dissociation energy from Armentrout
# red dots .. residual intensity (TIC -fragment - parent)
# black dots .. relative intensity of the fragment

# Ab-initio conversion ratios derivation
ev = 1.602176634e-19
mol = 6.02214076e23
farad = ev*mol
ev2kj = farad/1000
kj2ev = 1/ev2kj

# Energy-resolved CID reference dissociation energies are taken from:
#
# BH_HH, BH_HOMe, BH_MeMe:
# ========================
# https://pubs.acs.org/doi/pdf/10.1021/acs.analchem.9b02257
# Benzhydrylpyridinium Ions: A New Class of Thermometer Ions for the
# Characterization of Electrospray-Ionization Mass Spectrometers
#
# R. Rahrt, T. Auth, M. Demireva, P. B. Armentrout, K. Koszinowski, Anal. Chem.
# 2019, 91, 11703–11711.
# Took from the Abstract - the values are the same as the one in the Table2
#
# p-H, p-Me, p-OMe, p-NO2, p-NO2_alt:
# ===================================
# https://link.springer.com/content/pdf/10.1007/s13361-017-1693-0.pdf
# How Hot are Your Ions Really? A Threshold Collision-Induced Dissociation
# Study of Substituted Benzylpyridinium “Thermometer” Ions
#
# J. E. Carpenter, C. P. McNary, A. Furin, A. F. Sweeney, P. B. Armentrout, J.
# Am. Soc. Mass Spectrom. 2017, 28, 1876–1888.
# Took from the Table2, p-NO2_alt means the dissociation of the NO2
#
#
# Due to the argumentation used by Armenrout in the abovementioned paper (2017)
# we ommited the values presented in:
# https://www.sciencedirect.com/science/article/pii/S1387380616303608
# Experimental bond dissociation energies of benzylpyridinium thermometer ions
# determined by threshold-CID and RRKM modeling
#
# D. Gatineau, A. Memboeuf, A.  Milet, R. B. Cole, H. Dossmann, Y. Gimbert, D.
# Lesage, International Journal of Mass Spectrometry 2017, 417, 69–75.


# energies, uncertainty (eV) - from Armentrout experimental values
energs = {"p-H": (2.58, 0.15),
          "p-Me": (2.26, 0.13),
          "p-OMe": (1.93, 0.08),
          "p-NO2": (3.04, 0.12),
          "p-NO2_alt": (2.68, 0.13),
          "BH_HH": (1.79, 0.11),
          "BH_MeMe": (1.55, 0.13),
          "BH_HOMe": (1.37, 0.14)}

# masses to select - parent, fragment
ions = {"p-H": (170, 91),
        "p-Me": (184, 105),
        "p-OMe": (200, 121),
        "p-NO2": (215, 136),
        "p-NO2_alt": (215, 169),
        "BH_HH": (246, 167),
        "BH_MeMe": (274, 195),
        "BH_HOMe": (276, 197)}

# a bit generous peakwidth
peakwidth = 1.5

def boltzmann(x, bottom, top, xhalf, slope):
    """boltzmann sigmoid curve function"""
    y = bottom + (top - bottom) / (1 + np.exp((x-xhalf)/slope))
    return y

def fitline(x, a):
    return x*a


def get_tangentparams(parameters):
    """generate a, b for y = ax + b from sigomid fit parameters"""
    bottom, top, xhalf, slope = parameters[0]
    a = -(top - bottom) / (4*slope)
    yhalf = top + (bottom - top) / 2
    b = yhalf - a * xhalf
    return a, b


def get_r_sqrd(function, xdata, ydata, popt):
    """routine for getting fit probability"""
    residuals = ydata - function(xdata, *popt)
    ss_res = np.sum(residuals**2)
    ss_tot = np.sum((ydata-np.mean(ydata))**2)
    r_sqrd = 1 - (ss_res / ss_tot)
    return r_sqrd


figure = plt.figure(figsize=(8, 8), dpi=100, constrained_layout=True)
grid = figure.add_gridspec(int(np.ceil(len(energs)/2))+1, 2)


xAE, yAE, dates, yerrs = [], [], [], []
for n,i in enumerate(energs.keys()):
    print("processing %s..." % i)

    # data readout and processing
    filename = i+".RAW" if not i == "p-NO2_alt" else "p-NO2.RAW"
    if not os.path.isfile(filename):
        raise FileNotFoundError(
            errno.ENOENT, os.strerror(errno.ENOENT), filename)
    dates.append(os.stat(filename).st_mtime)
    matrix = load_raw(filename)[0]
    legends = load_params(filename)
    normenergs = np.asarray(legends[1]).T[4]
    times = matrix[0][0]
    assert len(times) == len(normenergs)
    masses = matrix[1]
    parentargs = np.where((masses > (ions[i][0]-peakwidth/2)) &
                          (masses < (ions[i][0]+peakwidth/2)))
    daughterargs = np.where((masses > (ions[i][1]-peakwidth/2)) &
                            (masses < (ions[i][1]+peakwidth/2)))
    #tic == zero -> bad scan
    tic = np.sum(matrix[2].T, axis=0)
    assert len(times) == len(tic)
    good = np.where(tic > 0)
    parentints = np.divide(np.sum(matrix[2].T[parentargs], axis=0), tic)[good]
    daughterints = np.divide(np.sum(matrix[2].T[daughterargs], axis=0), tic)[good]
    rest = 1 - parentints - daughterints
    
    # fitting
    x = normenergs[good]
    y = daughterints
    try:
       guess = [min(y), max(y), x[np.argmax(y>(max(y)/2))], 0.1]
       parameters = spopt.curve_fit(boltzmann, x, y, guess)
       print(parameters[0])
    except:
        print("optimization failed, giving up")
        raise Exception
 
    a, b = get_tangentparams(parameters)
    zerocross = -b/a
    yline = [0, max(y)]
    xline = [zerocross, (yline[1]-b)/a]
    rsqrd = get_r_sqrd(boltzmann, x, y, parameters[0])
    xfit = np.linspace(x[0],x[-1],500)
    yfit = boltzmann(xfit, *parameters[0])

    plot = figure.add_subplot(grid[int(n/2),n % 2], ylabel="$I_{frag}\ /\ \Sigma I$",
                              xlim=(min(x), max(x)), ylim=(-0.1,1.1))
    plot.spines['top'].set_visible(False)
    plot.spines['right'].set_visible(False)
    plot.plot(x, rest, '.', color = [1, 0.8, 0.8, 0.1])
    plot.plot(x, daughterints, '.', color = [0,0,0,0.1])
    plot.plot(xfit, yfit, color=[0,0,0,0.4])
    plot.plot(xline, yline, color=[0,0,0,0.7])
    plot.text(0.01, 1, 
              "{}\n$AE={:.2f}$\n$E_{{tcid}}={}\ eV$\n$r^2={:.4f}$".format(
        i, zerocross, energs[i][0], rsqrd), 
        transform=plot.axes.transAxes, va='top')

    xAE.append(zerocross)
    yAE.append(energs[i][0])
    yerrs.append(energs[i][1])
    print(zerocross)
    print("DONE! (%s)" % i)

finalfit = spopt.curve_fit(fitline, xAE, yAE)
k = finalfit[0][0]
rsq = get_r_sqrd(fitline, np.asarray(xAE), np.asarray(yAE), finalfit[0])

figure.suptitle("Calibration plots\n{} - {}".format(
    time.ctime(min(dates)), time.ctime(max(dates))))

print("Plotting the fitted values")
plot = figure.add_subplot(grid[-1,:], xlabel="$AE$",
                          ylabel="$E_{{tcid}}$")
plot.errorbar(xAE, yAE, yerr=yerrs, color = [0,0,0,0.4], linestyle='None' )
plot.plot(xAE, yAE, '.', color = 'black')
plot.plot((min(xAE), max(xAE)), (min(xAE)*k, max(xAE)*k), color = 'red')
for n,i in enumerate(energs.keys()):
    plot.annotate(i, (xAE[n]+0.1, yAE[n]), rotation=0, va='top')
plot.text(0.01, 1, "$k={:.4f}\ eV\ ({:.4f}\ kJ/mol),\ R^2 ={:.2f}$".format(
    k, k * farad/1000, rsq), transform=plot.axes.transAxes, va='top')
plt.savefig("doublecol.png")