Optimizing the Isolation Width in Orbitrap Instruments to ......Title Optimizing the Isolation Width...
1
Carmen Paschke 1 ; David Horn 2 ; Tabiwang N. Arrey 1 ; Rosa Rakownikow Jersie-Christensen 1 ; Martin Zeller 1 ; Romain Huguet 2 ; Pedro Navarro 1 , Bernard Delanghe 1 ; 1 Thermo Fisher Scientific, Bremen, Germany; 2 Thermo Fisher Scientific, San Jose, CA RESULTS Validation of the SEQUEST HT results To validate the identifications of the chimeric spectra, artificial chimeric spectra were created in silico by combining the mass peaks from known synthetic peptides from the Proteome Tools project (1). One to up to 10 of those spectra were merged into one and 1000 of each of those spectra were searched using SEQUEST HT. The validation was performed with a classical target decoy approach or using Percolator. With the known peptide sequences in each spectrum the real FDR was calculated by dividing the false positive identifications by all identifications (see figure 3). As expected, Percolator outperformsg the Target Decoy approach in number of identified peptides. The identification rate decreases with increasing multiplexing rate but is nevertheless superior to 85% using Percolator with FDR well below 1 %. Precision increases with greater isolation width; this is due to the fact that more peptides are identified. The accuracy slightly decreseswith MPS and increased isolation width, mainly due to the fact that the majority of additional proteins are of lower abundance (see Figures 10. and 11.) ABSTRACT Purpose: Maximizing the number of quantified proteins in Label Free Quantification experiments. Methods: Optimizing isolation width and using improved identification methods. Results: Optimized method produces 50% more protein groups. INTRODUCTION In recent years, the performance of the filtering quadrupole in Orbitrap instruments has been improved. Isolation widths down to 0.4 Th are available. For reporter based quantification, smaller widths should be used to avoid interference from the co-isolated peptides. For Label Free Quantification (LFQ) that might not be the case. Here we will investigate the optimal isolation width for maximizing the number of identified and quantified proteins and peptides, together with the best possible accuracy and precision in LFQ experiments on Thermo Scientific ™ Orbitrap Eclipse™ Tribrid™ and Thermo Scientific™ Orbitrap Exploris™ 480 mass spectrometers. MATERIALS AND METHODS Sample Preparation and Data Acquisition For the LFQ experiments different amount of yeast protein digest (Promega) was mixed in a constant HeLa protein digest (Pierce) background (200ng) to obtain a ratio of 2, 5 and 10. The samples were loaded onto Thermo Scientific™ EASY-Spray™ PepMap™ RSLC C18, 25 cm C18 column and separated with a 60 min gradient (8-30 % B [80% ACN in 0.1% FA] in 60 min, 5min to 50 %, another 5 min to 90 %B, 8 min at 90 % B). The eluting peptides were analyzed on the Orbitrap Exploris 480 mass spectrometer. The system was operated in a data dependent mode, selecting as many precursors as possible in 1 second cycle time. Isolation widths from 0.4 Th to 4 Th were used for MS2. For the LFQ data using High-Field Asymmetric Waveform Ion Mobility (FAIMS), the Thermo Scientific™ FAIMS Pro™ interface was connected to an Orbitrap Eclipse Tribrid mass spectrometer. A similar sample and chromatography was used as described above. The isolation width was set to 3 Th, MS2 fragments were measured in the Orbitrap analyzerr. Data Analysis Data analysis was performed using a beta version of Thermo Scientific™ Proteome Discoverer™ 2.4 software. To analyze the chimeric spectra, a new node, Precursor Detector, was developed. This node detects all precursor masses that are co-isolated with the targeted precursor and consequently fragmented together to produce the chimeric MS2 spectrum. Each precursor was then searched with that MS2 spectrum either using SEQUEST HT or MSPepSearch search engines. FASTA files used for SEQUEST HT: Homo sapiens (SwissProt TaxID=9606) (v2017-10-25) and Saccharomyces cerevisiae (SwissProt TaxID=4932_and_subtaxonomies) (v2017-10-25). The Human and Yeast spectral libraies used with MSPepSearch were predicted using PROSIT (2) with the following settings: sequence length 7 to 30, precursor charge 2 and 3, up to 2 missed cleavages, up to 2 M(ox) per peptide. All data were filtered at 1% FDR on psm, peptide and protein level. CONCLUSIONS Optimal isolation width for Label Free Quantification is around 3 to 4 Th. Up to 50% more protein groups can be identified searching for multiple peptides per spectrum using SEQUEST HT as well as MSPepSearch. Using FAIMS dramatically increases the accuracy and precision of quantified protein groups. FAIMS also increases the number of quantified protein groups REFERENCES 1. D. Zolg et al, Nature Methods, 14, 259 (2017): “Building ProteomeTools based on a complete synthetic human proteome”. 2. Gessulat & Schmidt et al. Nature Methods (2019): “ Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning”. ACKNOWLEDGEMENTS We would like to thank Dr Mathias Wilhelm and Siegfried Gessulat from the Technical University of Munich for supplying the PROSIT predicted spectral libraries for Human and Yeast. TRADEMARKS/LICENSING © 2019 Thermo Fisher Scientific Inc. All rights reserved. SEQUEST is a trademark of the University of Washington. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe the intellectual property rights of others. Optimizing the Isolation Width in Orbitrap Instruments to Maximize the Number of Label-Free Quantified Peptides and Proteins Figure 8. Precision and accuracy of the quantified Yeast protein groups for the different identification methods at isolation width 0.4 Th Figure 9. Precision and accuracy of the quantified Yeast protein groups for the different identification methods at isolation width 2 Th Figure 3. Number of identified multiplexed spectra versus multiplexing rate for SEQUEST HT Number of quantified and correctly (< 20% of expected ratio) quantified protein groups. MSPepSearch with MPS outperforms the rest Figure 5. Identified Protein Groups at 1% FDR for the different isolation widths Optimal isolation width for maximal peptide and protein identification All the samples measured with the different isolation widths have been analysed with SEQUEST HT and MSPepSearch, without and with searching for chimeric of multiple precursors (MPS). All validation was performed with Percolator. The number of identified protein groups increases with increasing isolation width, maximizing around 3 to 4 Th. MSPepSearch with MPS, in all cases, produces the most psms, peptide groups and protein groups (all at 1% FDR) increasing the number of protein groups by more then 50 % (figure 5.). Precision increases with increasing isolation width. This is due to the act that more peptides are identified (Figure 8. and 9.). MPS is slightly decreasing the precision due to the fact that more low abundant peptides and proteins are detected (Figure 10. and 11.). Accuracy for high abundant proteins is excellent for all data (Figure 7.) For lower abundant proteins (additional proteins identified with MPS) this is slightly worse (Figure 8., 9., 10. and 11.). 0 1000 2000 3000 4000 5000 6000 7000 8000 1 2 3 4 5 6 7 8 9 10 Target Decoy 1% FDR TRUE FALSE 0.0 0.3 0.3 0.5 0.3 0.4 0.3 0.5 0.3 0.6 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 1 2 3 4 5 6 7 8 9 10 Percolator 1% FDR TRUE FALSE 0.0 0.7 1.0 0.8 0.9 0.9 0.6 1.1 0.6 0.9 Actual FDR (%) Figure 1. Processing workflow Figure 2. Consensus workflow The x axis displays the number of multiplexed spectra per spectrum. SEQUEST HT shows excellent recovery rates, from 96% to 85% for 1 up to 10 multiplexed MS2 spectra with real FDR below 1%. As expected, Percolator outperforms Target Decoy FDR calculation. Figure 4. Number of identified multiplexed spectra versus multiplexing rate for MSPepSearch MSPepSearch showsexcellent recovery rates, from 96% to 88% for 1 up to 10 multiplexed MS2 spectra with real FDR below 1%. As expected, Percolator also outperforms Target Decoy FDR calculation. Validation of the MSPepSearch results The same approach as for SEQUEST HT (see above) has been used for MSPepSearch . 0 1000 2000 3000 4000 5000 6000 0.4 0.8 1.2 1.6 2 3 4 Protein Groups SequestHT SequestHT with MPS MSPEP MSPEP with MPS + 22% + 27% + 27% + 32% + 34% + 41% + 36% + 42% + 38% + 47% + 49% + 52% + 39% + 42% Figure 5 shows the number of identified protein groups in a mixture 200ng Hela + 25ng Yeast digest using different insolation width. Peptide identifications was done using SequestHT and MSPepsearch with and without multiple peptide search (MPS). Around 50 % more protein groups are identified with the chimeric search maximizing around 3 to 4 Th. Other samples are showing similar increases (data not shown). Precision and accuracy of the LFQ with and without MPS are shown in figures 6-9. Figure 6. Number of quantified Human protein groups at isolation width 2 Th for the different identification methods Figure 7. Precision and accuracy of the quantified Human protein groups for the different identification methods at isolation width 2 Th Figure 10. Protein ratio distribution at isolation width 0.4 Th for the different identification methods Figure 11. Protein ratio distribution at isolation width 2 Th for the different identification methods Using FAIMS Adding the FAIMS Pro interface in front of the Orbitrap Eclipse Tribrid dramatically improves the precision (figures 12 and 13) and accuracy of the quantified peptides and proteins. In addition, that also increases the number of Identified proteins (Figure 12. and Figure 13.) Figure 12. Number of quantified Yeast protein groups at isolation width 3 Th for the different identification methods Figure 13. Precision and accuracy of the quantified Yeast protein groups for the different identification methods at isolation width 3 Th and ratio of 5 Figure 14. Peptide ratio distribution at isolation width 3 Th for the different identification methods Figure 15. Protein ratio distribution at isolation width 3 Th for the different identification methods Adding FAIMS in front of the Orbitrap Eclipse Tribrid improves the accuracy and precision of the peptide and protein ratios . Accuracy of the ratios is excellent for all the isolation widths, however the precision is improving for larger isolation widths MPS is identifying and quantifying more low abundant peptides and proteins. Common between with and without MPS are marked in green, unique ones in organge. Using FAIMS improves greatly the accuracy and precision, especially for lower abundant proteins. FAIMS improves the number of identified and quantified protein groups. PO65545-EN0519S 0.0 0.8 0.9 0.5 0.8 1.1 0.6 0.7 0.8 0.9 Actual FDR (%) 0.0 0.8 1.0 0.9 0.9 0.9 1.0 1.0 0.8 0.9 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 1 2 3 4 5 6 7 8 9 10 Percolator 1% FDR TRUE FALSE 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 1 2 3 4 5 6 7 8 9 10 Target Decoy 1% FDR TRUE FALSE
Transcript of Optimizing the Isolation Width in Orbitrap Instruments to ......Title Optimizing the Isolation Width...
-
Carmen Paschke1; David Horn2; Tabiwang N. Arrey1; Rosa Rakownikow Jersie-Christensen1; Martin Zeller1; Romain Huguet2; Pedro Navarro1, Bernard Delanghe1; 1Thermo Fisher Scientific, Bremen, Germany; 2Thermo Fisher Scientific, San Jose, CA
RESULTS
Validation of the SEQUEST HT resultsTo validate the identifications of the chimeric spectra, artificial chimeric spectra were created in silico by combining the mass peaks from known synthetic peptides from the Proteome Tools project (1). One to up to 10 of those spectra were merged into one and 1000 of each of those spectra were searched using SEQUEST HT. The validation was performed with a classical target decoy approach or using Percolator. With the known peptide sequences in each spectrum the real FDR was calculated by dividing the false positive identifications by all identifications (see figure 3). As expected, Percolator outperformsg the Target Decoy approach in number of identified peptides. The identification rate decreases with increasing multiplexing rate but is nevertheless superior to 85% using Percolator with FDR well below 1 %.
Precision increases with greater isolation width; this is due to the fact that more peptides are identified. The accuracy slightly decreseswith MPS and increased isolation width, mainly due to the fact that the majority of additional proteins are of lower abundance (see Figures 10. and 11.)
ABSTRACT
Purpose: Maximizing the number of quantified proteins in Label Free Quantification experiments.
Methods: Optimizing isolation width and using improved identification methods.
Results: Optimized method produces 50% more protein groups.
INTRODUCTIONIn recent years, the performance of the filtering quadrupole in Orbitrap instruments has been improved. Isolation widths down to 0.4 Th are available. For reporter based quantification, smaller widths should be used to avoid interference from the co-isolated peptides. For Label Free Quantification (LFQ) that might not be the case. Here we will investigate the optimal isolation width for maximizing the number of identified and quantified proteins and peptides, together with the best possible accuracy and precision in LFQ experiments on Thermo Scientific™ Orbitrap Eclipse™ Tribrid™ and Thermo Scientific™ Orbitrap Exploris™ 480 mass spectrometers.
MATERIALS AND METHODSSample Preparation and Data Acquisition
For the LFQ experiments different amount of yeast protein digest (Promega) was mixed in a constant HeLa protein digest (Pierce) background (200ng) to obtain a ratio of 2, 5 and 10. The samples were loaded onto Thermo Scientific™ EASY-Spray™ PepMap™ RSLC C18, 25 cm C18 column and separated with a 60 min gradient (8-30 % B [80% ACN in 0.1% FA] in 60 min, 5min to 50 %, another 5 min to 90 %B, 8 min at 90 % B). The eluting peptides were analyzed on the Orbitrap Exploris 480 mass spectrometer. The system was operated in a data dependent mode, selecting as many precursors as possible in 1 second cycle time. Isolation widths from 0.4 Th to 4 Th were used for MS2.
For the LFQ data using High-Field Asymmetric Waveform Ion Mobility (FAIMS), the ThermoScientific™ FAIMS Pro™ interface was connected to an Orbitrap Eclipse Tribrid mass spectrometer. A similar sample and chromatography was used as described above. The isolation width was set to 3 Th, MS2 fragments were measured in the Orbitrap analyzerr.
Data AnalysisData analysis was performed using a beta version of Thermo Scientific™ Proteome Discoverer™ 2.4 software. To analyze the chimeric spectra, a new node, Precursor Detector, was developed. This node detects all precursor masses that are co-isolated with the targeted precursor and consequently fragmented together to produce the chimeric MS2 spectrum. Each precursor was then searched with that MS2 spectrum either using SEQUEST HT or MSPepSearch search engines. FASTA files used for SEQUEST HT: Homo sapiens (SwissProt TaxID=9606) (v2017-10-25) and Saccharomyces cerevisiae (SwissProt TaxID=4932_and_subtaxonomies) (v2017-10-25).
The Human and Yeast spectral libraies used with MSPepSearch were predicted using PROSIT (2) with the following settings: sequence length 7 to 30, precursor charge 2 and 3, up to 2 missed cleavages, up to 2 M(ox) per peptide.
All data were filtered at 1% FDR on psm, peptide and protein level.
CONCLUSIONS Optimal isolation width for Label Free Quantification is around 3 to 4 Th.
Up to 50% more protein groups can be identified searching for multiple peptides per spectrum using SEQUEST HT as well as MSPepSearch.
Using FAIMS dramatically increases the accuracy and precision of quantified protein groups. FAIMS also increases the number of quantified protein groups
REFERENCES1. D. Zolg et al, Nature Methods, 14, 259 (2017): “Building ProteomeTools based on a complete
synthetic human proteome”.
2. Gessulat & Schmidt et al. Nature Methods (2019): “ Prosit: proteome-wide prediction of peptide tandem mass spectra by deep learning”.
ACKNOWLEDGEMENTSWe would like to thank Dr Mathias Wilhelm and Siegfried Gessulat from the Technical University of Munich for supplying the PROSIT predicted spectral libraries for Human and Yeast.
TRADEMARKS/LICENSING© 2019 Thermo Fisher Scientific Inc. All rights reserved. SEQUEST is a trademark of the University of Washington. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries. This information is not intended to encourage use of these products in any manner that might infringe the intellectual property rights of others.
Optimizing the Isolation Width in Orbitrap Instruments to Maximize the Number of Label-Free Quantified Peptides and Proteins
Figure 8. Precision and accuracy of the quantified Yeast protein groups for the different identification methods at isolation width 0.4 Th
Figure 9. Precision and accuracy of the quantified Yeast protein groups for the different identification methods at isolation width 2 Th
Figure 3. Number of identified multiplexed spectra versus multiplexing rate for SEQUEST HT
Number of quantified and correctly (< 20% of expected ratio) quantified protein groups. MSPepSearch with MPS outperforms the rest
Figure 5. Identified Protein Groups at 1% FDR for the different isolation widths
Optimal isolation width for maximal peptide and protein identification All the samples measured with the different isolation widths have been analysed with SEQUEST HT and MSPepSearch, without and with searching for chimeric of multiple precursors (MPS). All
validation was performed with Percolator. The number of identified protein groups increases with increasing isolation width, maximizing around 3 to 4 Th. MSPepSearch with MPS, in all cases, produces the most psms, peptide groups and protein groups (all at 1% FDR) increasing the number of protein groups by more then 50 % (figure 5.). Precision increases with increasing isolation width. This is due to the act that more peptides are identified (Figure 8. and 9.). MPS is slightly decreasing the precision due to the fact that more low abundant peptides and proteins are detected (Figure 10. and 11.). Accuracy for high abundant proteins is excellent for all data (Figure 7.) For lower abundant proteins (additional proteins identified with MPS) this is slightly worse (Figure 8., 9., 10. and 11.).
0
1000
2000
3000
4000
5000
6000
7000
8000
1 2 3 4 5 6 7 8 9 10
Target Decoy 1% FDR
TRUE FALSE
0.0 0.3 0.3 0.5 0.3 0.4 0.3 0.5 0.3 0.6
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
1 2 3 4 5 6 7 8 9 10
Percolator 1% FDR
TRUE FALSE
0.0 0.7 1.0 0.8 0.9 0.9 0.6 1.1 0.6 0.9 Actual FDR (%)
Figure 1. Processing workflow Figure 2. Consensus workflow
The x axis displays the number of multiplexed spectra per spectrum. SEQUEST HT shows excellent recovery rates, from 96% to 85% for 1 up to 10 multiplexed MS2 spectra with real FDR below 1%. As expected, Percolator outperforms Target Decoy FDR calculation.
Figure 4. Number of identified multiplexed spectra versus multiplexing rate for MSPepSearch
MSPepSearch showsexcellent recovery rates, from 96% to 88% for 1 up to 10 multiplexed MS2 spectra with real FDR below 1%. As expected, Percolator also outperforms Target Decoy FDR calculation.
Validation of the MSPepSearch resultsThe same approach as for SEQUEST HT (see above) has been used for MSPepSearch .
0
1000
2000
3000
4000
5000
6000
0.4 0.8 1.2 1.6 2 3 4
Protein GroupsSequestHT SequestHT with MPS MSPEP MSPEP with MPS
+ 22%+ 27% + 27%
+ 32% + 34%+ 41%
+ 36%+ 42%
+ 38%+ 47% + 49%
+ 52%+ 39% + 42%
Figure 5 shows the number of identified protein groups in a mixture 200ng Hela + 25ng Yeast digest using different insolation width. Peptide identifications was done using SequestHT and MSPepsearch with and without multiple peptide search (MPS). Around 50 % more protein groups are identified with the chimeric search maximizing around 3 to 4 Th. Other samples are showing similar increases (data not shown).Precision and accuracy of the LFQ with and without MPS are shown in figures 6-9.
Figure 6. Number of quantified Human protein groups at isolation width 2 Th for the different identification methods
Figure 7. Precision and accuracy of the quantified Human protein groups for the different identification methods at isolation width 2 Th
Figure 10. Protein ratio distribution at isolation width 0.4 Th for the different identification methods
Figure 11. Protein ratio distribution at isolation width 2 Th for the different identification methods
Using FAIMSAdding the FAIMS Pro interface in front of the Orbitrap Eclipse Tribrid dramatically improves the precision (figures 12 and 13) and accuracy of the quantified peptides and proteins. In addition, that also increases the number of Identified proteins (Figure 12. and Figure 13.)
Figure 12. Number of quantified Yeast protein groups at isolation width 3 Th for the different identification methods
Figure 13. Precision and accuracy of the quantified Yeast protein groups for the different identification methods at isolation width 3 Th and ratio of 5
Figure 14. Peptide ratio distribution at isolation width 3 Th for the different identification methods
Figure 15. Protein ratio distribution at isolation width 3 Th for the different identification methods
Adding FAIMS in front of the Orbitrap Eclipse Tribrid improves the accuracy and precision of the peptide and protein ratios .
Accuracy of the ratios is excellent for all the isolation widths, however the precision is improving for larger isolation widths
MPS is identifying and quantifying more low abundant peptides and proteins. Common between with and without MPS are marked in green, unique ones in organge.
Using FAIMS improves greatly the accuracy and precision, especially for lower abundant proteins.
FAIMS improves the number of identified and quantified protein groups.
PO65545-EN0519S
0.0 0.8 0.9 0.5 0.8 1.1 0.6 0.7 0.8 0.9Actual FDR (%)0.0 0.8 1.0 0.9 0.9 0.9 1.0 1.0 0.8 0.9
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
10000
1 2 3 4 5 6 7 8 9 10
Percolator 1% FDR
TRUE FALSE
0
1000
2000
3000
4000
5000
6000
7000
8000
9000
1 2 3 4 5 6 7 8 9 10
Target Decoy 1% FDR
TRUE FALSE
Sheet2
spectrum cloning, no peaks are removedsubtraction of identified peaks befor the next search is done
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:10 ppm1997019971
Fragment match tolerance:0.02Da21948462194747
328701183289695
4377718743809179
5469127954653337
6556339565447505
7643351876263709
8731662886960976
98187768976241322
1089549661081991750
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1995019911
Fragment match tolerance:0.1Da2187610821847139
3272426432485491
43503469428731107
54284676530371948
649461036630082968
755531425730733891
860171935831704758
966752271929255978
10708528341030596860
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1990619844
Fragment match tolerance:0.5Da2180818821740252
32594394319671018
43241731418732102
537891191517663219
643331655616554307
747122267715145425
850932873813776455
955023453913027338
10556943401011538210
Copy & paste the data from the tables above here, to have them plotted
# PrecursorscorrectFALSE
19844
21740252
319671018
418732102
517663219
616554307
715145425
813776455
913027338
1011538210
12345678910984174019671873176616551514137713021153425210182102321943075425645573388210
Sheet3
merging 1 to 10 spectra from Proteome Tools library and searching the with SequestHT
spectrum cloning, no peaks are removedsubtraction of identified peaks before the next search is done
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Target Decoy Validator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1955251201101866261040010
2189821281416110.72153812398193151.2353706112
3277262432736421.032727105401431500.5133846718
43549160572891780.8434092757230731130.8800234673
5435124292401351410.953825646149331081940.862745098
65067387131451551910.9639521206313211423180.5313765182
75568745156342042650.6740351772454181945060.4460966543
86329799192672573321.1836142553781242037840.6640841173
964091286382412455020.693224325411261823911120.5583126551
1071101370473652965860.9103695309214184925814071.3261163735
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Percolator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1993002000.01993001000.0010070493
2192716054050.32165325837212220.1209921355
32766949732740.3328501811871120.6315789474
43696481020591310.5437176461382860.3497444175
54609601716941940.3545738319281111500.6122895255
654637323201112740.46534012933581862361.0861423221
7546465431615854240.37599518841752773681.2510425354
8710519246381344530.58672617836993275521.4719000892
976922989925906700.39722828453863558951.1898173769
10851534395541597430.610756040011711442812611.5079365079
Precursor match tolerance 10ppm, Fragment match tolerance 0.5 Da
# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
Percolator19335300300
2183312110281060.545553737
3257934156473030.2326483133
43253411325835650.7685213649
537521169611829180.2931769723
643319877339413190.7619487416
745662451321510918630.3285151117
84911264178268124670.5294237426
950773982742211630230.4333267678
1050314904382210038790.4372888094
Copy & paste the data from the tables above here, to have them plotted
# Precursorshighmediumlowwrong highwrong mediumwrong low
199300200
219271605405
3276694973274
4369648102059131
5460960171694194
65463732320111274
754646543161585424
871051924638134453
97692298992590670
1085153439554159743
# Precursors12345678910high993192727663696460954635464710576928515medium01694486073654192298343low009101723316469995wrong high25720162015382554wrong medium0403259941118513490159wrong low0574131194274424453670743
Target Decoy Validator !% FDR
TRUE12345678910955189827723549435150675568632964097110FALSE123456789100142728404534674165
Percolator 1% FDR
TRUE12345678910993192727663696460954635464710576928515FALSE1234567891025720162015382554
Sheet1
0.00.71.00.80.90.90.61.10.60.9Actual FDR (%)
ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!
0.0020140986908360.2594706798131810.2530730296456980.5411255411255410.3471468865263610.3660992128866920.2745241581259150.5348346235045740.3250130005200210.634174985320023
0.00.30.30.50.30.40.30.50.30.6
Sheet2
spectrum cloning, no peaks are removedsubtraction of identified peaks befor the next search is done
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:10 ppm1997019971
Fragment match tolerance:0.02Da21948462194747
328701183289695
4377718743809179
5469127954653337
6556339565447505
7643351876263709
8731662886960976
98187768976241322
1089549661081991750
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1995019911
Fragment match tolerance:0.1Da2187610821847139
3272426432485491
43503469428731107
54284676530371948
649461036630082968
755531425730733891
860171935831704758
966752271929255978
10708528341030596860
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1990619844
Fragment match tolerance:0.5Da2180818821740252
32594394319671018
43241731418732102
537891191517663219
643331655616554307
747122267715145425
850932873813776455
955023453913027338
10556943401011538210
Copy & paste the data from the tables above here, to have them plotted
# PrecursorscorrectFALSE
19844
21740252
319671018
418732102
517663219
616554307
715145425
813776455
913027338
1011538210
12345678910984174019671873176616551514137713021153425210182102321943075425645573388210
Sheet3
merging 1 to 10 spectra from Proteome Tools library and searching the with SequestHT
spectrum cloning, no peaks are removedsubtraction of identified peaks before the next search is done
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Target Decoy Validator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1955251201101866261040010
2189821281416110.72153812398193151.2353706112
3277262432736421.032727105401431500.5133846718
43549160572891780.8434092757230731130.8800234673
5435124292401351410.953825646149331081940.862745098
65067387131451551910.9639521206313211423180.5313765182
75568745156342042650.6740351772454181945060.4460966543
86329799192672573321.1836142553781242037840.6640841173
964091286382412455020.693224325411261823911120.5583126551
1071101370473652965860.9103695309214184925814071.3261163735
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Percolator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1993002000.01993001000.0010070493
2192716054050.32165325837212220.1209921355
32766949732740.3328501811871120.6315789474
43696481020591310.5437176461382860.3497444175
54609601716941940.3545738319281111500.6122895255
654637323201112740.46534012933581862361.0861423221
7546465431615854240.37599518841752773681.2510425354
8710519246381344530.58672617836993275521.4719000892
976922989925906700.39722828453863558951.1898173769
10851534395541597430.610756040011711442812611.5079365079
Precursor match tolerance 10ppm, Fragment match tolerance 0.5 Da
# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
Percolator19335300300
2183312110281060.545553737
3257934156473030.2326483133
43253411325835650.7685213649
537521169611829180.2931769723
643319877339413190.7619487416
745662451321510918630.3285151117
84911264178268124670.5294237426
950773982742211630230.4333267678
1050314904382210038790.4372888094
Copy & paste the data from the tables above here, to have them plotted
# Precursorshighmediumlowwrong highwrong mediumwrong low
199300200
219271605405
3276694973274
4369648102059131
5460960171694194
65463732320111274
754646543161585424
871051924638134453
97692298992590670
1085153439554159743
# Precursors12345678910high993192727663696460954635464710576928515medium01694486073654192298343low009101723316469995wrong high25720162015382554wrong medium0403259941118513490159wrong low0574131194274424453670743
Target Decoy Validator !% FDR
TRUE12345678910955189827723549435150675568632964097110FALSE123456789100142728404534674165
Percolator 1% FDR
TRUE12345678910993192727663696460954635464710576928515FALSE1234567891025720162015382554
Sheet1
0.00.71.00.80.90.90.61.10.60.9Actual FDR (%)
ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!
0.0020140986908360.2594706798131810.2530730296456980.5411255411255410.3471468865263610.3660992128866920.2745241581259150.5348346235045740.3250130005200210.634174985320023
0.00.30.30.50.30.40.30.50.30.6
Sheet2
spectrum cloning, no peaks are removedsubtraction of identified peaks befor the next search is done
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:10 ppm1997019971
Fragment match tolerance:0.02Da21948462194747
328701183289695
4377718743809179
5469127954653337
6556339565447505
7643351876263709
8731662886960976
98187768976241322
1089549661081991750
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1995019911
Fragment match tolerance:0.1Da2187610821847139
3272426432485491
43503469428731107
54284676530371948
649461036630082968
755531425730733891
860171935831704758
966752271929255978
10708528341030596860
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1990619844
Fragment match tolerance:0.5Da2180818821740252
32594394319671018
43241731418732102
537891191517663219
643331655616554307
747122267715145425
850932873813776455
955023453913027338
10556943401011538210
Copy & paste the data from the tables above here, to have them plotted
# PrecursorscorrectFALSE
19844
21740252
319671018
418732102
517663219
616554307
715145425
813776455
913027338
1011538210
12345678910984174019671873176616551514137713021153425210182102321943075425645573388210
Sheet3
merging 1 to 10 spectra from Proteome Tools library and searching the with SequestHT
spectrum cloning, no peaks are removedsubtraction of identified peaks before the next search is done
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Target Decoy Validator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1955251201101866261040010
2189821281416110.72153812398193151.2353706112
3277262432736421.032727105401431500.5133846718
43549160572891780.8434092757230731130.8800234673
5435124292401351410.953825646149331081940.862745098
65067387131451551910.9639521206313211423180.5313765182
75568745156342042650.6740351772454181945060.4460966543
86329799192672573321.1836142553781242037840.6640841173
964091286382412455020.693224325411261823911120.5583126551
1071101370473652965860.9103695309214184925814071.3261163735
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Percolator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1993002000.01993001000.0010070493
2192716054050.32165325837212220.1209921355
32766949732740.3328501811871120.6315789474
43696481020591310.5437176461382860.3497444175
54609601716941940.3545738319281111500.6122895255
654637323201112740.46534012933581862361.0861423221
7546465431615854240.37599518841752773681.2510425354
8710519246381344530.58672617836993275521.4719000892
976922989925906700.39722828453863558951.1898173769
10851534395541597430.610756040011711442812611.5079365079
Precursor match tolerance 10ppm, Fragment match tolerance 0.5 Da
# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
Percolator19335300300
2183312110281060.545553737
3257934156473030.2326483133
43253411325835650.7685213649
537521169611829180.2931769723
643319877339413190.7619487416
745662451321510918630.3285151117
84911264178268124670.5294237426
950773982742211630230.4333267678
1050314904382210038790.4372888094
Copy & paste the data from the tables above here, to have them plotted
# Precursorshighmediumlowwrong highwrong mediumwrong low
199300200
219271605405
3276694973274
4369648102059131
5460960171694194
65463732320111274
754646543161585424
871051924638134453
97692298992590670
1085153439554159743
# Precursors12345678910high993192727663696460954635464710576928515medium01694486073654192298343low009101723316469995wrong high25720162015382554wrong medium0403259941118513490159wrong low0574131194274424453670743
Target Decoy Validator !% FDR
TRUE12345678910955189827723549435150675568632964097110FALSE123456789100142728404534674165
Percolator 1% FDR
TRUE12345678910993192727663696460954635464710576928515FALSE1234567891025720162015382554
Sheet1
0.00.71.00.80.90.90.61.10.60.9Actual FDR (%)
ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!
0.0020140986908360.2594706798131810.2530730296456980.5411255411255410.3471468865263610.3660992128866920.2745241581259150.5348346235045740.3250130005200210.634174985320023
0.00.30.30.50.30.40.30.50.30.6
Sheet2
spectrum cloning, no peaks are removedsubtraction of identified peaks befor the next search is done
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:10 ppm1997019971
Fragment match tolerance:0.02Da21948462194747
328701183289695
4377718743809179
5469127954653337
6556339565447505
7643351876263709
8731662886960976
98187768976241322
1089549661081991750
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1995019911
Fragment match tolerance:0.1Da2187610821847139
3272426432485491
43503469428731107
54284676530371948
649461036630082968
755531425730733891
860171935831704758
966752271929255978
10708528341030596860
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1990619844
Fragment match tolerance:0.5Da2180818821740252
32594394319671018
43241731418732102
537891191517663219
643331655616554307
747122267715145425
850932873813776455
955023453913027338
10556943401011538210
Copy & paste the data from the tables above here, to have them plotted
# PrecursorscorrectFALSE
19844
21740252
319671018
418732102
517663219
616554307
715145425
813776455
913027338
1011538210
12345678910984174019671873176616551514137713021153425210182102321943075425645573388210
Sheet3
merging 1 to 10 spectra from Proteome Tools library and searching the with SequestHT
spectrum cloning, no peaks are removedsubtraction of identified peaks before the next search is done
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Target Decoy Validator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1955251201101866261040010
2189821281416110.72153812398193151.2353706112
3277262432736421.032727105401431500.5133846718
43549160572891780.8434092757230731130.8800234673
5435124292401351410.953825646149331081940.862745098
65067387131451551910.9639521206313211423180.5313765182
75568745156342042650.6740351772454181945060.4460966543
86329799192672573321.1836142553781242037840.6640841173
964091286382412455020.693224325411261823911120.5583126551
1071101370473652965860.9103695309214184925814071.3261163735
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Percolator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1993002000.01993001000.0010070493
2192716054050.32165325837212220.1209921355
32766949732740.3328501811871120.6315789474
43696481020591310.5437176461382860.3497444175
54609601716941940.3545738319281111500.6122895255
654637323201112740.46534012933581862361.0861423221
7546465431615854240.37599518841752773681.2510425354
8710519246381344530.58672617836993275521.4719000892
976922989925906700.39722828453863558951.1898173769
10851534395541597430.610756040011711442812611.5079365079
Precursor match tolerance 10ppm, Fragment match tolerance 0.5 Da
# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
Percolator19335300300
2183312110281060.545553737
3257934156473030.2326483133
43253411325835650.7685213649
537521169611829180.2931769723
643319877339413190.7619487416
745662451321510918630.3285151117
84911264178268124670.5294237426
950773982742211630230.4333267678
1050314904382210038790.4372888094
Copy & paste the data from the tables above here, to have them plotted
# Precursorshighmediumlowwrong highwrong mediumwrong low
199300200
219271605405
3276694973274
4369648102059131
5460960171694194
65463732320111274
754646543161585424
871051924638134453
97692298992590670
1085153439554159743
# Precursors12345678910high993192727663696460954635464710576928515medium01694486073654192298343low009101723316469995wrong high25720162015382554wrong medium0403259941118513490159wrong low0574131194274424453670743
Target Decoy Validator !% FDR
TRUE12345678910955189827723549435150675568632964097110FALSE123456789100142728404534674165
Percolator 1% FDR
TRUE12345678910993192727663696460954635464710576928515FALSE1234567891025720162015382554
Sheet1
0.00.71.00.80.90.90.61.10.60.9Actual FDR (%)
ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!
0.0020140986908360.2594706798131810.2530730296456980.5411255411255410.3471468865263610.3660992128866920.2745241581259150.5348346235045740.3250130005200210.634174985320023
0.00.80.90.50.81.10.60.70.80.9
Sheet2
spectrum cloning, no peaks are removedsubtraction of identified peaks befor the next search is done
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:10 ppm1997019971
Fragment match tolerance:0.02Da21948462194747
328701183289695
4377718743809179
5469127954653337
6556339565447505
7643351876263709
8731662886960976
98187768976241322
1089549661081991750
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1995019911
Fragment match tolerance:0.1Da2187610821847139
3272426432485491
43503469428731107
54284676530371948
649461036630082968
755531425730733891
860171935831704758
966752271929255978
10708528341030596860
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1990619844
Fragment match tolerance:0.5Da2180818821740252
32594394319671018
43241731418732102
537891191517663219
643331655616554307
747122267715145425
850932873813776455
955023453913027338
10556943401011538210
Copy & paste the data from the tables above here, to have them plotted
# PrecursorscorrectFALSE
19844
21740252
319671018
418732102
517663219
616554307
715145425
813776455
913027338
1011538210
12345678910984174019671873176616551514137713021153425210182102321943075425645573388210
Sheet3
merging 1 to 10 spectra from Proteome Tools library and searching the with SequestHT
spectrum cloning, no peaks are removedsubtraction of identified peaks before the next search is done
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Target Decoy Validator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1955251201101866261040010
2189821281416110.72153812398193151.2353706112
3277262432736421.032727105401431500.5133846718
43549160572891780.8434092757230731130.8800234673
5435124292401351410.953825646149331081940.862745098
65067387131451551910.9639521206313211423180.5313765182
75568745156342042650.6740351772454181945060.4460966543
86329799192672573321.1836142553781242037840.6640841173
964091286382412455020.693224325411261823911120.5583126551
1071101370473652965860.9103695309214184925814071.3261163735
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Percolator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1993002000.01993001000.0010070493
2192716054050.32165325837212220.1209921355
32766949732740.3328501811871120.6315789474
43696481020591310.5437176461382860.3497444175
54609601716941940.3545738319281111500.6122895255
654637323201112740.46534012933581862361.0861423221
7546465431615854240.37599518841752773681.2510425354
8710519246381344530.58672617836993275521.4719000892
976922989925906700.39722828453863558951.1898173769
10851534395541597430.610756040011711442812611.5079365079
Precursor match tolerance 10ppm, Fragment match tolerance 0.5 Da
# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
Percolator19335300300
2183312110281060.545553737
3257934156473030.2326483133
43253411325835650.7685213649
537521169611829180.2931769723
643319877339413190.7619487416
745662451321510918630.3285151117
84911264178268124670.5294237426
950773982742211630230.4333267678
1050314904382210038790.4372888094
Copy & paste the data from the tables above here, to have them plotted
# Precursorshighmediumlowwrong highwrong mediumwrong low
199300200
219271605405
3276694973274
4369648102059131
5460960171694194
65463732320111274
754646543161585424
871051924638134453
97692298992590670
1085153439554159743
# Precursors12345678910high993192727663696460954635464710576928515medium01694486073654192298343low009101723316469995wrong high25720162015382554wrong medium0403259941118513490159wrong low0574131194274424453670743
Target Decoy Validator !% FDR
TRUE12345678910955189827723549435150675568632964097110FALSE123456789100142728404534674165
Percolator 1% FDR
TRUE12345678910993192727663696460954635464710576928515FALSE1234567891025720162015382554
Sheet1
0.00.71.00.80.90.90.61.10.60.9Actual FDR (%)
ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!
0.0020140986908360.2594706798131810.2530730296456980.5411255411255410.3471468865263610.3660992128866920.2745241581259150.5348346235045740.3250130005200210.634174985320023
0.00.30.30.50.30.40.30.50.30.6
Sheet2
spectrum cloning, no peaks are removedsubtraction of identified peaks befor the next search is done
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:10 ppm1997019971
Fragment match tolerance:0.02Da21948462194747
328701183289695
4377718743809179
5469127954653337
6556339565447505
7643351876263709
8731662886960976
98187768976241322
1089549661081991750
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1995019911
Fragment match tolerance:0.1Da2187610821847139
3272426432485491
43503469428731107
54284676530371948
649461036630082968
755531425730733891
860171935831704758
966752271929255978
10708528341030596860
# Precursorscorrectwrong# Precursorscorrectwrong
Precursor match tolerance:20 ppm1990619844
Fragment match tolerance:0.5Da2180818821740252
32594394319671018
43241731418732102
537891191517663219
643331655616554307
747122267715145425
850932873813776455
955023453913027338
10556943401011538210
Copy & paste the data from the tables above here, to have them plotted
# PrecursorscorrectFALSE
19844
21740252
319671018
418732102
517663219
616554307
715145425
813776455
913027338
1011538210
12345678910984174019671873176616551514137713021153425210182102321943075425645573388210
Sheet3
merging 1 to 10 spectra from Proteome Tools library and searching the with SequestHT
spectrum cloning, no peaks are removedsubtraction of identified peaks before the next search is done
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Target Decoy Validator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1955251201101866261040010
2189821281416110.72153812398193151.2353706112
3277262432736421.032727105401431500.5133846718
43549160572891780.8434092757230731130.8800234673
5435124292401351410.953825646149331081940.862745098
65067387131451551910.9639521206313211423180.5313765182
75568745156342042650.6740351772454181945060.4460966543
86329799192672573321.1836142553781242037840.6640841173
964091286382412455020.693224325411261823911120.5583126551
1071101370473652965860.9103695309214184925814071.3261163735
Precursor match tolerance 10ppm, Fragment match tolerance 0.02 DaPrecursor match tolerance 10ppm, Fragment match tolerance 0.02 Da
Percolator
# PrecursorsTRUEmediumlowFALSEwrong mediumwrong lowactual FDR high /%# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
1993002000.01993001000.0010070493
2192716054050.32165325837212220.1209921355
32766949732740.3328501811871120.6315789474
43696481020591310.5437176461382860.3497444175
54609601716941940.3545738319281111500.6122895255
654637323201112740.46534012933581862361.0861423221
7546465431615854240.37599518841752773681.2510425354
8710519246381344530.58672617836993275521.4719000892
976922989925906700.39722828453863558951.1898173769
10851534395541597430.610756040011711442812611.5079365079
Precursor match tolerance 10ppm, Fragment match tolerance 0.5 Da
# Precursorshighmediumlowwrong highwrong mediumwrong lowactual FDR high /%
Percolator19335300300
2183312110281060.545553737
3257934156473030.2326483133
43253411325835650.7685213649
537521169611829180.2931769723
643319877339413190.7619487416
745662451321510918630.3285151117
84911264178268124670.5294237426
950773982742211630230.4333267678
1050314904382210038790.4372888094
Copy & paste the data from the tables above here, to have them plotted
# Precursorshighmediumlowwrong highwrong mediumwrong low
199300200
219271605405
3276694973274
4369648102059131
5460960171694194
65463732320111274
754646543161585424
871051924638134453
97692298992590670
1085153439554159743
# Precursors12345678910high993192727663696460954635464710576928515medium01694486073654192298343low009101723316469995wrong high25720162015382554wrong medium0403259941118513490159wrong low0574131194274424453670743
Target Decoy Validator !% FDR
TRUE12345678910955189827723549435150675568632964097110FALSE123456789100142728404534674165
Percolator 1% FDR
TRUE12345678910993192727663696460954635464710576928515FALSE1234567891025720162015382554
Sheet1
0.00.71.00.80.90.90.61.10.60.9Actual FDR (%)
ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!ERROR:#DIV/0!
0.0020140986908360.2594706798131810.2530730296456980.5411255411255410.3471468865263610.3660992128866920.2745241581259150.5348346235045740.3250130005200210.634174985320023
0.00.81.00.90.90.91.01.00.80.9
Slide Number 1
![[width=1.7cm]Figures/kthengrgb.jpg [1em]Optimizing ...](https://static.fdocuments.net/doc/165x107/618510271b2b851a323298fa/width17cmfigures-1emoptimizing-.jpg)
