From 35b97de57b16a85324614e3708dbb03a7cf2fe74 Mon Sep 17 00:00:00 2001 From: kilian Date: Wed, 12 Aug 2026 14:48:54 +0200 Subject: [PATCH] Dateien nach "ILP/Vanilla" hochladen --- ILP/Vanilla/nmrSimilarityVanillien.py | 387 ++++++++++---------------- 1 file changed, 140 insertions(+), 247 deletions(-) diff --git a/ILP/Vanilla/nmrSimilarityVanillien.py b/ILP/Vanilla/nmrSimilarityVanillien.py index 6ca6439..da998eb 100644 --- a/ILP/Vanilla/nmrSimilarityVanillien.py +++ b/ILP/Vanilla/nmrSimilarityVanillien.py @@ -1,244 +1,177 @@ -#Binning mostly for broader peaks? -#0.66 für H und 8.4 für C bei anderen TMS Werten -#Gute 13C Ergebnisse für alles +11 ppm: Im Vergeich mit Coffein haben alle disubstituierten bei wenigen Hohen Werten falsche Zuordnung, bei mono und nicht substituierten sogar keine Falsche zuornung (0.1 bis 5 mit 0.1 Schritten) -#+11 nicht universell, aber 9 bis 13 bei allen sweet spot - import math import numpy as np -#Xanthine -HXANTHINE = { - 1: ([7.96], [1]), - 2: ([9.45], [1]), - 3: ([7.725], [1]), - 4: ([7.625], [1]), +["CINNAMICACID", "PCOUMARICACID", "MCOUMARICACID", "BENZALDEHYD", "CAFFEICACID", "3HYDROXYBENZALDEHYD", "4HYDROXYBENZALDEHYD", "34DIHYDROXYBENZALDEHYD"] +#H noch machen + +likelihood = [] +#Cinnamicacid +HCINNAMICACID = { + } -CXANTHINE = { - 1: ([159.40], [1]), - 2: ([164.01], [1]), - 3: ([120.94], [1]), - 4: ([161.24], [1]), - 5: ([146.98], [1]), +CCINNAMICACID = { + 1: ([171.56], [1]), + 2: ([120.33], [1]), + 3: ([151.41], [1]), + 4: ([140.47], [1]), + 5: ([134.20], [2]), + 6: ([133.48], [2]), + 7: ([135.91], [1]), } -#1-Methylxanthine -H1XANTHINE = { - 1: ([7.93], [1]), - 2: ([9.45], [1]), - 3: ([4.05], [3]), - 4: ([7.91], [1]), +#p-Coumaricacid +HPCOUMARICACID = { + } -C1XANTHINE = { - 1: ([161.50], [1]), - 2: ([166.28], [1]), - 3: ([120.65], [1]), - 4: ([158.74], [1]), - 5: ([146.25], [1]), - 6: ([38.55], [1]), +CPCOUMARICACID = { + 1: ([171.61], [1]), + 2: ([117.18], [1]), + 3: ([149.67], [1]), + 4: ([132.76], [1]), + 5: ([135.75], [2]), + 6: ([118.65], [2]), + 7: ([166.09], [1]), } -#3-Methylxanthine -H3XANTHINE = { - 1: ([4.15], [3]), - 2: ([7.73], [1]), - 3: ([7.99], [1]), - 4: ([9.49], [1]), +#m-Coumaricacid +HMCOUMARICACID = { + } -C3XANTHINE = { - 1: ([161.83], [1]), - 2: ([163.37], [1]), - 3: ([121.24], [1]), - 4: ([163.15], [1]), - 5: ([146.49], [1]), - 6: ([39.71], [1]), +CMCOUMARICACID = { + 1: ([172.55], [1]), + 2: ([118.60], [1]), + 3: ([153.69], [1]), + 4: ([141.29], [1]), + 5: ([121.09], [1]), + 6: ([134.46], [1]), + 7: ([121.09], [1]), + 8: ([164.15], [1]), + 9: ([122.36], [1]), } -#7-Methylxanthine -H7XANTHINE = { - 1: ([7.55], [1]), - 2: ([4.47], [3]), - 3: ([7.72], [1]), - 4: ([7.655], [1]), +#Benzaldehyd +HBENZALDEHYD = { + } -C7XANTHINE= { - 1: ([159.50], [1]), - 2: ([165.47], [1]), - 3: ([122.15], [1]), - 4: ([162.31], [1]), - 5: ([151.55], [1]), - 6: ([45.06], [1]), +CBENZALDEHYD= { + 1: ([195.09], [1]), + 2: ([141.47], [1]), + 3: ([135.86], [2]), + 4: ([133.45], [2]), + 5: ([139.15], [1]), } -#Theophylline -H13XANTHINE = { - 1: ([4.03], [3]), - 2: ([7.98], [1]), - 3: ([4.19], [3]), - 4: ([9.49], [1]), +#Caffeicacid +HCAFFEICACID = { + } -C13XANTHINE = { - 1: ([163.77], [1]), - 2: ([165.26], [1]), - 3: ([120.73], [1]), - 4: ([160.99], [1]), - 5: ([145.80], [1]), - 6: ([40.42], [1]), - 7: ([37.60], [1]), +CCAFFEICACID = { + 1: ([171.66], [1]), + 2: ([116.16], [1]), + 3: ([149.88], [1]), + 4: ([131.94], [1]), + 5: ([132.99], [1]), + 6: ([118.15], [1]), + 7: ([155.79], [1]), + 8: ([149.28], [1]), + 9: ([110.95], [1]), } -#Paraxanthine -H17XANTHINE = { - 1: ([4.50], [3]), - 2: ([7.70], [1]), - 3: ([3.98], [3]), - 4: ([7.82], [1]), +#3-Hydroxybenzaldehyd +H3HYDROXYBENZALDEHYD = { + } -C17XANTHINE = { - 1: ([161.41], [1]), - 2: ([167.17], [1]), - 3: ([121.81], [1]), - 4: ([160.18], [1]), - 5: ([151.09], [1]), - 6: ([45.17], [1]), - 7: ([36.96], [1]), +C3HYDROXYBENZALDEHYD = { + 1: ([164.39], [1]), + 2: ([122.95], [1]), + 3: ([133.96], [1]), + 4: ([123.33], [1]), + 5: ([143.41], [1]), + 6: ([195.16], [1]), + 7: ([124.61], [1]), } -CPARAXANTHINE = { - 1: ([26.7], [1]), - 2: ([32.9], [1]), - 3: ([151.1], [1]), - 4: ([106.5], [1]), - 5: ([147.4], [1]), - 6: ([155.3], [1]), - 7: ([143.0], [1]), +#4-Hydroxybenzaldehyd +H4HYDROXYBENZALDEHYD = { + +} +C4HYDROXYBENZALDEHYD = { + 1: ([191.53], [1]), + 2: ([135.08], [1]), + 3: ([137.69], [2]), + 4: ([118.46], [2]), + 5: ([168.28], [1]), } -#Theobromine -H37XANTHINE = { - 1: ([4.49], [3]), - 2: ([7.75], [1]), - 3: ([4.11], [3]), - 4: ([7.65], [1]), +#3,4-Dihydroxybenzaldehyd +H34DIHYDROXYBENZALDEHYD = { + } -C37XANTHINE = { - 1: ([161.76], [1]), - 2: ([164.86], [1]), - 3: ([122.51], [1]), - 4: ([164.29], [1]), - 5: ([151.10], [1]), - 6: ([39.33], [1]), - 7: ([45.07], [1]), +C34DIHYDROXYBENZALDEHYD = { + 1: ([158.08], [1]), + 2: ([117.79], [1]), + 3: ([133.40], [1]), + 4: ([134.66], [1]), + 5: ([191.41], [1]), + 6: ([113.10], [1]), + 7: ([149.32], [1]), } -#Caffeine -H137XANTHINE = { - 1: ([7.73], [1]), - 2: ([4.15], [3]), - 3: ([4.52], [3]), - 4: ([4.01], [3]), -} -C137XANTHINE = { - 1: ([163.66], [1]), - 2: ([166.66], [1]), - 3: ([122.03], [1]), - 4: ([162.23], [1]), - 5: ([150.50], [1]), - 6: ([40.09], [1]), - 7: ([45.23], [1]), - 8: ([37.17], [1]), -} - -CCAFFEINE = { - 1: ([155.7], [1]), #166 - 2: ([148.8], [1]), #159 - 3: ([107.7], [1]), #118 - 4: ([152.2], [1]), #163 - 5: ([143.0], [1]), #154 - 6: ([27.2], [1]), #38 - 7: ([29.1], [1]), #40 - 8: ([32.9], [1]), #44 -} - -CCAFFEINEADJUSTED = { - 1: ([166.7], [1]), #166 - 2: ([159.8], [1]), #159 - 3: ([118.7], [1]), #118 - 4: ([163.2], [1]), #163 - 5: ([154.0], [1]), #154 - 6: ([38.2], [1]), #38 - 7: ([40.1], [1]), #40 - 8: ([44.9], [1]), #44 -} -CCAFFEINE2 = { - 1: ([27.7], [1]), - 2: ([29.3], [1]), - 3: ([33.1], [1]), - 4: ([151.0], [1]), - 5: ([148.1], [1]), - 6: ([106.6], [1]), - 7: ([154.5], [1]), - 8: ([142.8], [1]), -} - -C137XANTHINEADJUSTED = { - 1: ([155.62], [1]), - 2: ([158.29], [1]), - 3: ([113.66], [1]), - 4: ([153.86], [1]), - 5: ([142.13], [1]), - 6: ([31.72], [1]), - 7: ([36.86], [1]), - 8: ([28.8], [1]), -} - -#Experimental 7-Methylxanthine nmr -#Secundary source 11.52, 3.81 +#Experimental p-Coumaricacid HNMR1= { - 1: ([10.85], [1]), - 2: ([11.50], [1]), - 3: ([3.82], [3]), - 4: ([7.88], [1]), + 1: ([12.13], [1]), + 2: ([7.49], [2]), + 3: ([6.79], [2]), + 4: ([9.96], [1]), + 5: ([6.29], [1]), + 6: ([7.52], [1]), } CNMR1= { - 1: ([155.85], [1]), - 2: ([151.35], [1]), - 3: ([149.30], [1]), - 4: ([143.01], [1]), - 5: ([106.90], [1]), - 6: ([33.03], [1]), + 1: ([125.36], [1]), + 2: ([130.17], [2]), + 3: ([115.83], [2]), + 4: ([159.67], [1]), + 5: ([168.05], [1]), + 6: ([115.41], [1]), + 7: ([144.27], [1]), } -#Experimental Theobromine nmr +#Experimental 4-Hydroxybenzaldehyd HNMR2 = { - 1: ([11.10], [1]), - 2: ([3.33], [3]), - 3: ([3.84], [3]), - 4: ([7.97], [1]), + 1: ([7.05], [2]), + 2: ([7.58], [2]), + 3: ([8.44], [1]), + 4: ([10.15], [1]), } +#No Intesities mentioned CNMR2 = { - 1: ([154.9], [1]), - 2: ([149.8], [1]), - 3: ([107.1], [1]), - 4: ([151.0], [1]), - 5: ([142.8], [1]), - 6: ([29.3], [1]), - 7: ([33.9], [1]), + 1: ([116.5], [2]), + 2: ([130.3], [1]), + 3: ([132.7], [2]), + 4: ([163.8], [1]), + 5: ([191.0], [1]), +} + +#Experimental 3,4-Dihydroxybenzaldehyd +HNMR3 = { + 1: ([7.44], [1]), + 2: ([7.42], [1]), + 3: ([7.00], [1]), } -#Combination of methyl group and base purine rings from two papers CNMR3 = { - 1: ([154.9], [1]), - 2: ([150.0], [1]), - 3: ([108.1], [1]), - 4: ([153.1], [1]), - 5: ([142.6], [1]), - 6: ([29.3], [1]), - 7: ([33.9], [1]), + 1: ([124.59], [1]), + 2: ([115.21], [1]), + 3: ([145.44], [1]), + 4: ([151.27], [1]), + 5: ([130.13], [1]), + 6: ([114.22], [1]), + 7: ([190.26], [1]), } def overlap(listref, listnew): @@ -294,65 +227,25 @@ def correction(spectra, corretionppm): return newspectra def main(): - spectrumref = CNMR2 - #spectra = [HXANTHINE, H1XANTHINE, H3XANTHINE, H7XANTHINE, H1XANTHINE, H17XANTHINE, H37XANTHINE, H137XANTHINE] - spectra = [CXANTHINE, C1XANTHINE, C3XANTHINE, C7XANTHINE, C1XANTHINE, C17XANTHINE, C37XANTHINE, C137XANTHINE] - spectranames = ["XANTHINE", "1XANTHINE", "3XANTHINE", "7XANTHINE", "1XANTHINE", "17XANTHINE", "37XANTHINE", "137XANTHINE"] + spectrumref = CNMR3 + #1H-NMR Spectra ignoriert, da meiste H sauer, da an N gebunden + #spectra = [HCINNAMICACID, HPCOUMARICACID, HMCOUMARICACID, HBENZALDEHYD, HCAFFEICACID, H3HYDROXYBENZALDEHYD, H4HYDROXYBENZALDEHYD, H34DIHYDROXYBENZALDEHYD] + spectra = [CCINNAMICACID, CPCOUMARICACID, CMCOUMARICACID, CBENZALDEHYD, CCAFFEICACID, C3HYDROXYBENZALDEHYD, C4HYDROXYBENZALDEHYD, C34DIHYDROXYBENZALDEHYD] + spectranames = ["CINNAMICACID", "PCOUMARICACID", "MCOUMARICACID", "BENZALDEHYD", "CAFFEICACID", "3HYDROXYBENZALDEHYD", "4HYDROXYBENZALDEHYD", "34DIHYDROXYBENZALDEHYD"] likelihood = [] for spectrumtrue in spectra: - #errorlist = {} - #errorlist = [] similaritybycorrection = [] - #Paper Chemical reviews Carbons bound to Heavy atoms (TMS) to high -> this could be reason for to high values. - correctionvalues = [9.37] #for C tested np.arange(8.0, 16.1, 0.1) range(8, 12) 8.4, 8.37, 11, 9.4 (for CNMR3), 9.87 (true for all ref, 8.37 + 1.5 for the precision), 9.37 (good for first, ok for second, third because only 7 better/equal but for first much higher) np.arange(6.13, 9.86, 0.01) (only for first), for H 0.66 (not good), np.arange(0.4, 1.0, 0.01), 0.6 for first, second never first either 17 or caf higher np.arange(0.51, 0.82, 0.01) good measure + #Paper Chemical reviews Carbons bound to Heavy atoms (TMS) to high -> this could be reason for too high values. + correctionvalues = [2.63] #np.arange(0, 1.51, 0.01) #for C tested np.arange(-0.37, 7.64, 0.1) 0, 2.63, 1 (for CNMR3), 1.5 (true for all ref, 8.37 + 1.5 for the precision), 1 (good for first, ok for second, third because only 7 better/equal but for first much higher) np.arange(-1.5, 1.49, 0.01) (only for first), for H 0 (not good), np.arange(-0.26, 0.34, 0.01), -0.06 for first, second never first either 17 or caf higher np.arange(-0.15, 0.16, 0.01) good measure for correctionvalue in correctionvalues: spectrumrefcorrected = correction(spectrumref, correctionvalue) #CCAFFEINE 11 (klappt hier sehr gut) CCAFFEINE2 12 CPARAXANTHINE 10 CNMR1 9, 10 o 11 (sehr gut) CNMR2 10 o 11 - #Likelihood by number of higher similarity than all others. - '''error = 0 - total = 0 - for spectrumfalse in spectra: - positive = 0 - negative = 0 - bad_binwidth = [] - for i in np.arange(0.1, 2.6, 0.1): #successfull at max 3.9, but max 1.7 is lowest where nmr3 correctly classified, 1.6 increases likelihood of 3,7 over 1,7 even with 8.4 correction - truesimilarity = similarity_nmr(spectrumtrue, spectrumrefcorrected, i) - falsesimilarity = similarity_nmr(spectrumfalse, spectrumrefcorrected, i) - #print(truesimilarity) - #print(falsesimilarity) - if(truesimilarity - falsesimilarity < 0 or truesimilarity == 0): - negative += 1 - bad_binwidth.append(i) - else: - positive += 1 - total += 1 - #print(f'Wrong similarity result: {negative} and Right similarity result: {positive}') - #print(bad_binwidth) - error += negative - #errorlist[correctionvalue] = error - errorlist.append(error) - print(min(range(len(errorlist)), key=errorlist.__getitem__)) - likelihood.append(round((total - min(errorlist))/total, 2))''' - #Likelihood by mean similarity - #This method demonstrates the same problems as the other likelihood method similaritylist = [] - binwidthlist = np.arange(0.1, 3.9, 0.1) #np.arange(0.1, 3.9, 0.1) + binwidthlist = np.arange(0.1, 3.9, 0.1) for i in binwidthlist: similaritylist.append(similarity_nmr(spectrumtrue, spectrumrefcorrected, i)) similaritymean = sum(similaritylist) / len(similaritylist) similaritybycorrection.append(similaritymean) - name = spectranames[spectra.index(spectrumtrue)] - correctionindex = max(range(len(similaritybycorrection)), key=similaritybycorrection.__getitem__) - print(f'{name}: {correctionindex} = {correctionvalues[correctionindex]}') - #Maybe not the best but a mean instead? likelihood.append(round(sum(similaritybycorrection)/len(similaritybycorrection), 2)) - #likelihood.append(round(max(similaritybycorrection), 2)) print(likelihood) - '''for i in np.arange(0.01, 0.07, 0.01): - print(f'Increment i: {i}') - print(similarity_nmr(HNMR1, HNMR2, i)) - print(similarity_nmr(H1XANTHINE, HNMR1, i)) - print(similarity_nmr(H3XANTHINE, HNMR1, i)) - print(similarity_nmr(H7XANTHINE, HNMR1, i)) - ''' if __name__ == "__main__": main() \ No newline at end of file