Research

MRMaid, the Web-based Tool for Designing

Multiple Reaction Monitoring (MRM) Transitions*
Jennifer A. Mead‡, Luca Bianco‡, Vanessa Ottone‡, Chris Barton§, Richard G. Kay§,
Kathryn S. Lilley¶, Nicholas J. Bond¶, and Conrad Bessant‡储

Multiple reaction monitoring (MRM) of peptides uses tan- available over the internet as an executable web-based
dem mass spectrometry to quantify selected proteins of service at www.mrmaid.info. Molecular & Cellular Pro-
interest, such as those previously identified in differential teomics 8:696 –705, 2009.
studies. Using this technique, the specificity of precursor
to product transitions is harnessed for quantitative anal-
ysis of multiple proteins in a single sample. The design of Multiple reaction monitoring (MRM)1 is a mass spectro-
transitions is critical for the success of MRM experiments, metry (MS)-based technique for monitoring the absolute
but predicting signal intensity of peptides and fragmenta- amount of specific proteins of interest. MS and MS/MS are
tion patterns ab initio is challenging given existing meth- performed while reverse phase (RP) HPLC separation is in
ods. The tool presented here, MRMaid (pronounced “mer- progress. Each tryptic peptide is analyzed by selection on the
maid”) offers a novel alternative for rapid design of MRM basis of mass using a quadrupole MS (Q1). Once separated,
transitions for the proteomics researcher. The program
it undergoes fragmentation in the collision cell, generating
uses a combination of knowledge of the properties of
product ions exclusive to the precursor, which are selectively
optimal MRM transitions taken from expert practitioners
and literature with MS/MS evidence derived from interro- monitored by a third quadrupole (Q3). This two-stage filtering
gation of a database of peptide identifications and their process allows chemical background to be overcome by im-
associated mass spectra. The tool also predicts retention proving signal to noise ratio and permits several transitions to
time using a published model, allowing ordering of tran- be monitored quickly. MRM can be performed as a quanti-
sition candidates. By exploiting available knowledge and tative method by spiking the sample with a known quantity
resources to generate the most reliable transitions, this of labeled synthetic peptide, which is identical in sequence
approach negates the need for theoretical prediction of to the expected target peptide (1). The observed m/z ratio of
fragmentation and the need to undertake prior “discov- a peptide and its corresponding product ion m/z ratio are
ery” MS studies. MRMaid is a modular tool built around referred to as a MRM “transition.” Consequently to monitor
the Genome Annotating Proteomic Pipeline framework, a protein of interest, it must be known in advance which
providing a web-based solution with both descriptive and
transition is most suitable. In simple protein mixtures, a
graphical visualizations of transitions. Predicted transi-
single transition may be sufficient to monitor a particular
tion candidates are ranked based on a novel transition
scoring system, and users may filter the results by select- protein of interest, but in complex samples, such as serum,
ing optional stringency criteria, such as omitting fre- multiple transitions are generally required because of noise
quently modified residues, constraining the length of pep- and proteins of very high abundance interfering with the
tides, or omitting missed cleavages. Comparison with signal (2, 3).
published transitions showed that MRMaid successfully The growth in popularity of MRM has come about because
predicted the peptide and product ion pairs in the majority biologists require techniques to measure protein regulation
of cases with appropriate retention time estimates. As the across many targets simultaneously and in a quantitative
data content of the Genome Annotating Proteomic Pipe- manner. MRM has the capability to meet this need (2–14) with
line repository increases, the coverage and reliability of quantitation reported down to femtomole concentration (8,
MRMaid are set to increase further. MRMaid is freely 14). Indeed for discovery and validation of novel biomarkers,
MRM has proven to be a successful method (4, 5, 10, 11) and
compared with the alternatives, such as ELISAs, has the
From the ‡Bioinformatics Group, Cranfield Health, Building 63, advantage of being cost-effective, quicker to design, and
Cranfield University, College Road, Cranfield, Bedfordshire MK43
0AL, United Kingdom, ¶Cambridge Centre for Proteomics, Depart-
suitable for multiplexed analysis (13). Increased throughput is
ment of Biochemistry, University of Cambridge, Tennis Court Road,
1
Cambridge CB2 1QR, United Kingdom, and §Quotient Bioresearch The abbreviations used are: MRM, multiple reaction monitoring;
Ltd., Newmarket Road, Fordham, Cambridgeshire CB7 5WW, United GAPP, The Genome Annotating Proteomic Pipeline; MIDAS, MRM-
Kingdom initiated detection and sequencing; RT, retention time; SRM, select-
Received, April 30, 2008, and in revised form, November 10, 2008 ed/single reaction monitoring; TIQAM, Targeted Identification for
Published, MCP Papers in Press, November 15, 2008, DOI Quantitative Analysis by MRM; TS, transition score; RP, reverse
10.1074/mcp.M800192-MCP200 phase.

696 Molecular & Cellular Proteomics 8.4 © 2009 by The American Society for Biochemistry and Molecular Biology, Inc.
This paper is available on line at http://www.mcponline.org
 Designing MRM Transitions Using MRMaid

also possible with MRM because of direct coupling of sepa- TIQAM is designed to be installed locally, requiring setup of
ration (via HPLC) to MS and, in some cases, the ability to a local database.
avoid extensive sample preparation before analysis (2, 3). In addition to the above, “research-grade” software proto-
Furthermore MRM requires a high level of ion separation but not types for designing MRM transitions have also been de-
necessarily high resolution, meaning that the instrumentation is scribed (8). In this particular example, software performed an
potentially affordable compared with the alternatives because in silico digest, computed the m/z of precursor/fragment ions,
lower resolution MS is generally less expensive to run. and filtered results on amino acid content. Multiple possible
The major challenge for MRM is deciding which peptides to transitions were generated. The problem with this approach is
monitor because each protein has multiple tryptic cleavage that an “appropriate” fragment ion was chosen; in MRMaid no
sites. In many cases, a lab-based discovery phase must be such theoretical prediction is needed because empirical data
performed prior to MRM, allowing observation of peptide is used.
performance in MS/MS to direct the decision (4, 12, 15). This Finally, some groups apply diverse bioinformatics re-
empirical method is effective but consumes time and labora- sources to transition design rather than using an integrated
tory resources and does not utilize the growing library of software solution (as those described above). Notably Ander-
existing public proteomics data. Moreover the level of detail son and Hunter (10) predicted frequently observed peptides
given of the method used for transition design is extremely using The Global Proteomics Machine’s (GPM) proteotypic
variable. We believe, therefore, that there is a need for a library (18) and used techniques to predict retention time (RT).
transparent, automated method to assist the design of tran- This piecemeal approach is difficult to reproduce, and the
sitions and to support the inevitable growth of the MRM algorithms applied are often not released to the public.
approach in the biological community. In summary, there are limited options available for MRM
Commercial, vendor-specific software packages for MRM practitioners requiring transition design support without in-
transition design are available, including for example MIDASTM volving acquisition of MS/MS data for the prediction process.
(MRM-initiated detection and sequencing) Workflow Designer In this study, a publicly accessible program, called MRMaid, is
software (Applied Biosystems, Foster City, CA), which calcu- presented, that provides transition design support to assist
lates theoretical peptides and corresponding transitions and the expansion in use of the MRM method. It is the only
then builds the MIDAS acquisition method (7) whereby a software option where transitions are based not only on prior
coupled Q-TRAP (Applied Biosystems) iteratively cycles knowledge of MRM and a published RT model but also on an
through scans and selects suitable peptides. Thermo Scien- increasingly large body of diverse MS/MS evidence, negating
tific’s (Waltham, MA) contribution has been presented to user the need for data generation for ranked predictions to be
groups but was not available for purchase at the time of made (for proteins present in the database).
writing. It takes data from SIEVETM , the automated label-free MRMaid is also vendor-independent, providing integrated
differential expression software, computes ions generated, transition design support for any group worldwide. Further-
relates those back to the sequence, and then imposes filters more the MRMaid algorithm is totally transparent as it is fully
similar to those in MRMaid. described in detail here and accompanying on-line documen-
A very recent addition to the field is Targeted Identifica- tation. It is web-based with an intuitive user interface and
tion for Quantitative Analysis by MRM (TIQAM) (16). TIQAM does not require computer expertise to use, avoiding down-
mines the PeptideAtlas repository (17) for peptide candi- load and complicated processes for setup locally. Results can
dates based on the number of previous observations. When be downloaded as a spreadsheet for storage and analysis.
no data is available, all theoretically possible “proteotypic” MRMaid also supports diverse MS instruments and RP chro-
peptides are computed using physiochemical properties matography conditions.
alone. “Proteotypic” means a peptide that uniquely maps to
EXPERIMENTAL PROCEDURES
a protein, so when it is detected it confirms the presence of
the protein. Based on user preferences, transitions are gen- MRMaid Overview
erated by TIQAM for the peptide list. In a final step, MS/MS The MRMaid method for transition design relies on the combination
for discovery of MRM transitions is performed, and the of two sources of information: first on prior knowledge of the kind of
results are mapped onto the list of transition candidates. precursor peptides that generally perform better in MS/MS and sec-
ond on mining the data in a proteomics data repository, Genome
TIQAM fundamentally differs from MRMaid because it re- Annotating Proteomic Pipeline (GAPP) (19), to determine which pre-
quires the user to experimentally acquire MS/MS data to cursor and corresponding fragment ions have appeared regularly for
isolate suitable candidates from the list of possible transi- the given protein of interest. The hydrophobicity and RP RT of each
tions, whereas MRMaid is able to indicate which peptides in peptide candidate are calculated so suitable transitions may be se-
the short list are most suitable using its novel transition lected and ordered. Finally as with all approaches for transition de-
sign, the short list of the best transition candidates must be validated
scoring algorithm. Other differences between MRMaid and using a suitable MS instrument.
TIQAM include the differing source of MS/MS reference The GAPP database is populated with data from public resources
data and the fact that MRMaid is a web-based tool whereas and data submitted directly by users. MS/MS peak lists and corre-

Molecular & Cellular Proteomics 8.4 697

Designing MRM Transitions Using MRMaid

sponding metadata (such as cleavage agent, mass tolerances, spe- Users may define the proportion of times a peptide has been
cies, and number of missed cleavages) are submitted via the website, observed for the target protein in the GAPP database. In this way, the
and GAPP performs identification of peptides using an X!Tandem frequency of peptide identifications in the repository may be used as
search (20). Protein identifications are then inferred using the ad- a measure of transition reliability for the protein. As a default, peptides
vanced average peptide score algorithm (21). with internal cleavage sites (namely peptides with Lys or Arg not
Peptide candidates for MRM are mined from the database by followed by Pro) are omitted to prevent selection of peptides that may
applying the principle of proteotypicality (22) whereby peptides that be irregularly cleaved. This is a necessary feature because the effi-
map unambiguously to a single protein are first extracted from the ciency of trypsin is known to be only (approximately) 70% (24, 25).
database. In GAPP, proteotypic uniqueness in the proteome is deter- Another important consideration for users intending to perform
mined for each candidate peptide at the average peptide score stage MRM is whether MS/MS evidence is available for their particular MS
of processing so further filtering is not required to retrieve unique instrument. Each instrument is known to have a different set of
peptides for MRMaid. preferred proteotypic peptides (22); therefore, to account for this
To design a MRM experiment, several SRM transitions are moni- phenomenon, MS instrument information in the GAPP database may
tored in a single assay. For this purpose, MRMaid allows comparison be incorporated as a filter in the querying process. A drop-down
of individual SRM transition candidates using estimated RT and hy- menu to select the type of instrument is provided on the home page.
drophobicity values; this way transitions may be selected to avoid Protein expression can vary significantly between tissue types. To
co-elution of peptides. The proposed peptide candidates can be try to account for this, the type of biological sample can also be
compared by downloading the page of MRMaid results and compar- specified as a filter in the search for transitions. In serum, this is
ing RT and transition score (TS) values. This process is explained particularly useful because its levels of protein expression and sample
below and ultimately allows users to design the optimal bespoke complexity can present unique challenges for transition design. The
MRM experiment for their specific target proteins. ability to choose transitions based upon experiments performed on
To indicate reliability of the transitions, metrics of reproducibility the same type of sample increases the likelihood of selection of
are calculated for the candidate product ions. An indication of successful candidates.
reproducibility is required because GAPP is a public system and, as In addition to constraining experiment-specific factors, information
such, accepts data from any source if the format and metadata on the sequence of the peptide may also be used in the filtering
requirements are met. Reliability is ensured in two discrete ways. process. For example, Asn and Gln may be deselected because these
First, peptide precursor (and subsequent fragment) candidates are residues can be deamidated resulting in fragment ion m/z irregulari-
presented to the user with the number of times they have been ties and problems with reproducibility. This is a problem because the
observed in GAPP for the protein of interest. Second, the individual m/z of fragment ions must be as consistent and hence reliable as
product ions are assessed in terms of signal intensity reproducibil- possible. Opting to omit peptides containing Gln or Glu at the N
ity: an average of signal intensity for the relevant m/z peaks is terminus is also possible in MRMaid because these residues can
calculated across all applicable experiments as well as the variance spontaneously cyclize to form pyroglutamate. Likewise Pro is an
and standard deviation values. These descriptive statistics indicate
important residue to consider when designing transitions. Peptides
the reproducibility of the fragment ions for a given precursor and
containing Pro may be considered favorable because they generally
hence point to the number of times one would expect to have to run
produce MS/MS peaks of high intensity. This is because the three-
the MRM experiment to observe a good result for the transition.
dimensional structure of proline promotes fragmentation, often pro-
ducing a single greater abundance fragment ion that “swamps” the
Software Implementation remainder of the tandem mass spectrum. However, this single signal
MRMaid (Fig. 1) is a program written in Perl and PHP that interro- may not be sufficient to identify the protein with adequate specificity,
gates the GAPP database for suitable transitions for a given protein particularly in complex samples. For this reason, users may opt to
sequence. The protein target is input by the user as a database acces- omit or allow candidate peptides containing Pro. The location of Pro
sion number, such as an Ensembl, Swiss-Prot, or International Protein in the peptide primary sequence is also important. Pro that is adjacent
Index number. Many database accession numbers are supported to the C terminus (P1) or in the second position from the C terminus
thanks to integration of the European Bioinformatics Institute’s free (P2) is generally not desirable in MRM. This is because a very short,
Protein Identifier Cross-Referencing Tool (23). The steps in the algorithm nonspecific product y-ion is produced that is unsuitable for monitor-
include the following. (a) Proteotypic peptide candidates for the protein ing. Users may, therefore, opt to omit P1- and P2-containing peptides
are retrieved. (b) Peptides are filtered by criteria defined by the user. (c) from their results.
MS/MS data for the protein are retrieved, descriptive statistics are Peptide sequence can also affect the probability of covalent mod-
calculated, and y- and b-ions are assigned for each peptide using the ification. Met and Cys are often modified residues, so if they are
mass tolerance window provided by the original data submitter. (d) A TS omitted it makes it possible to constrain the candidates to those
value is derived for each peptide and is used to rank the transitions. (e) where mass should not vary. Naturally if all possible amino acids that
Finally RT and hydrophobicity are computed for each peptide. The fall into this category were omitted, the filter would be too strict;
following sections explain the MRMaid work flow (Fig. 1) in more detail. however, by negatively weighting these residues in the TS calculation
Filtering for Optimal Transition Candidates—Users may choose it provides a more suitable method (see the next section for more
from a series of filtering options to constrain the prediction of transi- detail). Typical modifications like carbamidomethylation, however, are
tions (Table I). These filters take the form of a list of check boxes, worth considering in peptides for monitoring as there are examples in
drop-down menus, and text boxes on the home page. They can be the literature, such as for apolipoprotein A-II (2), where they have been
found below the box where the user enters the accession number for present in transitions, so for this users may choose a filter in this list
the protein target. The filters were chosen through targeted question- to omit them.
ing of experts in the MRM approach and the use of prototypes for live A further filter option is peptide length. Users may constrain this
demonstration. The reasoning behind the inclusion of these criteria is because very short peptides (⬍7– 8 residues) are unlikely to be
described below and is summarized in Table I. All peptides that pass unique, and peptides longer than ⬃20 –25 residues are unsuitable for
the relevant filtering steps enter the transition scoring phase, which is MRM because they may exceed the acceptable mass range. For all
necessary to rank the candidates. MRMaid searches, mass range for the peptide (MS mode) is restricted

698 Molecular & Cellular Proteomics 8.4

 Designing MRM Transitions Using MRMaid

FIG. 1. The process of MRM transition design showing the criteria used to derive transition score. TS is computed using MS/MS data
in the GAPP database unless there are insufficient data available for the protein of interest. TS is applied to rank the resulting transition
predictions. x is a value that can be adjusted according to the user’s requirements. DB, database; info., information.

Molecular & Cellular Proteomics 8.4 699

Designing MRM Transitions Using MRMaid

TABLE I
Optional Boolean filters that may be used to constrain the search for MRM transitions using MRMaid
Filter criterion Description
Peptide observation redundancy Peptide must have been seen in x% of all observations of the query protein
in GAPP database
Internal cleavage site Peptides with Lys or Arg not followed by Pro
Instrument type Constrains to MS/MS data retrieved for the instrument setup selected
Omit Asn and Gln Asn and Gln can be deamidated resulting in m/z reproducibility issues
Peptide length Short peptides (⬍7 or 8 amino acids) are unlikely to be unique to the target,
and very long peptides will be out of mass range
Omit Met and Cys Often covalently modified, affecting m/z
Omit Gln and Glu Can spontaneously cyclize to form pyroglutamate
Accept only Pro-containing peptides Produces a very high abundance peak, suitable when a single product ion is
sufficient, such as in low complexity samples
Omit Pro at any position Selected when multiple transitions per target are required, such as in high
complexity samples like serum
Omit P1 (Pro adjacent to the C terminus) Can produce nonspecific product y-ion
Omit P2 (Pro second position from C terminus) Can produce nonspecific product y-ion
m/z cut off (user specifies a value, x) Selects only fragment ion m/z ⬎x where x is a percentage of the precursor m/z

to 500 –1600 m/z. This range is routinely used for monitoring tryptic experts and would have made it impossible to judge relative favor-
peptides, so peptide candidates beyond this range are omitted by ability of candidate transitions. Therefore, TS is calculated as the
default. weighted sum of several key characteristics of the spectral data (each
b- and y-ions are common fragment ions produced in tandem MS. denoted as a letter: q, c, r, s, p and n), giving a quantitative measure
To restrict transitions to peptides that produce suitable y- or b-ions of expected performance in MRM (Table II). The relative weights
for monitoring, two approaches are used in MRMaid. First, y-ions applied (50, 8, 7, 6, 1 and 1, respectively) were determined by com-
(shown in red) and b-ions (in blue), with the charge states, are high- bining authors’ experience with researching the literature and discus-
lighted in the resulting spectrum graphic and table of results. This is sions with practicing MRM experts. Note that all values are normal-
achieved by dynamic computation using the accepted rules of frag- ized to a range between 0 and 1 such that each coefficient has
mentation (26) and where the theoretical m/z values (for monoisotopic consistent scale.
and average mass as required) are compared with those observed in The MS/MS suitability score (q) positively weights precursor pep-
the underlying MS/MS data. The mass tolerance window is applied to tides that demonstrate a suitable profile of y-ions for MRM. A suitable
ensure the resolution of the data is accounted for in the assignment; y-ion profile is one where there are several y-ions at the high m/z
for this we use the range specific to each experiment in GAPP and not range of the MS/MS spectrum. y-ions (with a 1⫹ charge state) were
the maximum overall relevant observations in the search. In an ex- chosen for this because these generally have higher m/z than b-ions,
tremely small number of cases, this window prevents unambiguous so they are a suitable indicator to determine the quality of a spectrum
assignment of a peak to a single b- or y-ion. In this case, the ion is not for MRM. They are also the ions that are routinely used as demon-
counted as b- or y-ion and is instead grouped into the “other ion type” strated, for example, in Anderson and Hunter’s study (10) where
category (shown in black). virtually all experimentally confirmed transitions were y-ions: see Ta-
The second approach to ensure MRMaid suggests suitable frag- ble III where selected transitions from this study are used to validate
ment peaks is by using m/z cut off, an approach that is demon- MRMaid performance. Furthermore b-ions are less suitable than y-
strated in the literature (3, 10). This facilitates selection of fragments ions because they are susceptible to cyclization, which can result in
with m/z greater than the precursor m/z. If desired, only MS/MS fragment ions of unexpected sequence (27). Therefore, the q coeffi-
fragment masses that are x% or more of the mass of the precursor cient (and hence overall TS value) is designed to favor spectra show-
will be recommended by MRMaid where x is specified by the user. ing evidence of multiple y-ions and is achieved by considering three
For example, if x is 100, then only fragment ions having a mass key factors for each precursor peptide sequence that passes the
higher than the precursor ion will be selected. This ensures that initial filtering stage for the protein of interest: first, the highest m/z
fragments in the higher end of the m/z spectrum will be considered value of the MS/MS spectrum that is a y-ion with 1⫹ charge ((m/
and accounts for the effect of the mass filtering step that is applied z)max); second, the number of peaks of the spectrum that are y-ions
between MS and MS/MS modes. This approach has the knock-on with 1⫹ charge (npeaks); and third, the standard deviation of m/z
effect of increasing the chances of producing a more reliable and values of peaks of the spectrum that are y-ions with 1⫹ charge
specific fragment candidate because the specificity of a transition (stdpeaks). Standard deviation is calculated in the usual way (see
is greatly increased when looking at a product ion with m/z higher Equation 1) where (m/z)peaks is the individual m/z value of a y-ion with
than the precursor ion. Note that there is no limit to the number of 1⫹ charge in the MS/MS spectrum and (m/z)peaks is the mean m/z
fragment ions that can be suggested to the user; only the underly- value of all the individual y-ions with 1⫹ charge for the given spectrum
ing data restricts this. for the peptide in question.

冑冘
Transition Scoring—The MRMaid TS provides a quantitative meas-
ure of predicted performance in MRM to candidate transitions that are 共共m/z兲peaks ⫺ 共m/z兲peaks兲2
retrieved by mining the GAPP database and by aggregating the stdpeaks ⫽ (Eq. 1)
npeaks ⫺ 1
results. Implementation of a series of Boolean filtering steps (as in
the filters described above) was one available option for this part of The rationale for using the three factors above is the following: if it
the process. However, this would not reflect the approach taken by is possible for a peptide to fragment into n different y-ions, then the

700 Molecular & Cellular Proteomics 8.4

 Designing MRM Transitions Using MRMaid

TABLE II
Derivation of TS coefficients
TS is used to rank the predicted transitions in MRMaid. It is calculated as the sum of the coefficients, each relating to efficiency in MRM.
Each coefficient is derived by multiplying the value by the weighting.
Transition score coefficient Description Relative weighting
MS/MS suitability (q) Assesses MS/MS spectrum for suitability in MRM 50
by assessing m/z values of y-ions
Peptide coverage (c) Favors fragment ions that represent a greater 8
proportion of the precursor peptide sequence
Mass range (r) Constrains acceptable MS/MS mode mass range 7
Precursor charge state (s) Doubly and triply charge peptides are favored 6
Positively weighted residue content (p) His, Lys, and Arg 1
Negatively weighted residue content (n) Tyr, Ser, Thr, Cys, and Met 1

MS/MS metric for suitability in MRM should favor spectra with evi- played as b- and y-ion labels on the peaks so the user can select
dence of larger y-ions (those having higher m/z values) as well as a singly charged peaks for monitoring.
higher number of y-ions in total. Because in general we cannot expect Weighting of specific residue content (coefficients p and n) allows
to observe all theoretically possible y-ions for a given peptide, we more refined ranking to be performed on the candidate output list.
must instead favor spectra with a smaller standard deviation of m/z Positively charged residues have been demonstrated to increase
values belonging to identified y-ions. fragmentation and hence “flyability”, so they are positively weighted
The three elements above are combined to assign higher scores to (22). Tyr, Ser, and Thr, like Cys and Met, may be post-translationally
spectra carrying evidence of the heaviest y-ion as well as the greater modified, so they are negatively weighted.
number of y-ions. If two or more spectra are equal according to these Transition Scoring in the Absence of MS/MS Evidence—In addition
first two criteria, the one identifying a higher number of heavier y-ions to the above, TS may also be calculated in the absence of MS/MS
gets a higher score. Thus, given a spectrum for a peptide sequence in data. In these situations, it is computed with q omitted, and the user
the GAPP database, its interim score kr (r denoting “real”) is calcu- is informed on the results page. This calculation involves digestion of
lated as follows. the protein and application of user-defined filters as in the usual
MRMaid mode. Clearly candidates predicted in the absence of MS/MS
共m/z兲max 䡠 npeaks evidence are less robust, because predicted peptides cannot be sta-
kr ⫽ (Eq. 2)
冑1 ⫹ stdpeaks tistically measured for reliability across experimental data sets, and less
comprehensive, because candidate MS/MS fragment ions cannot be
After this, using the peptide sequence, a theoretical interim score kt suggested. However, an indication of which peptides should be moni-
(t for “theoretical”) is similarly calculated but this time considering tored is welcome functionality because the alternative is deciding on
the total complement of theoretical y-ions (with 1⫹ charge state) for transitions manually, which would be tedious for large proteins and
the sequence using the rules of CID fragmentation (26). To finish, the particularly problematic when many proteins are to be analyzed.
MS/MS suitability score (q) is computed as a ratio between the real
and theoretical interim scores; thus, q is normalized to a scale be- RESULTS
tween 0 and 1 as is the case of all other coefficients in TS.
Downloading Results from MRMaid—In the final output,
kr MRMaid provides a tabulated list of ranked transition candi-
q⫽ (Eq. 3)
kt dates, which are intended for validation using a MS instru-
In summary, q provides a quantitative scale of MS/MS suitability to
ment. This validation is necessary because it is not possible to
each peptide based on the MS/MS spectral evidence and is weighted select a single in silico-derived transition without experimental
most heavily because it is quantifying actual experimental data. confirmation as shown by existing software options, such as
Peptide coverage (c) refers to the proportion of the target peptide MIDAS Workflow Designer. The table of results may be ex-
that is represented by the product ions for the transition. Product ions ported as tab-separated values for import into a spreadsheet
that have greater m/z are preferable because they represent a greater
proportion of the original peptide in the spectrum and therefore
package for local analysis and archiving. Also as described
increase the specificity of the transition. Peptides with fragments above, schematic spectra are displayed to highlight the y- and
within the mass range for MS/MS mode (r) of 500 –1600 m/z are b-ions suggested for monitoring.
positively weighted for reasons mentioned earlier. Retention Time—Elution time data are commonly used in
Most tryptic peptides are doubly charged cations with one charge combination with MS/MS to support transition design for
originating from the primary N terminus and one from the Lys or Arg
residue side chain at the C terminus. Precursor charge state (s) is
MRM experiments (3, 12, 13). RT is used in the discovery
important for MRM because doubly or triply charged precursor pep- phase to decide which peptides to monitor on the basis of
tides are favored because of the mass filtering stage. Doubly and peak separation, and once transitions are chosen, it allows
triply charged precursors, therefore, achieve a coefficient higher than transition ordering (12) and provides confirmation that the
singly charged peptides, such as those derived from unspecific pro- peptide species expected is the one actually being monitored
tein cleavage. Charge state for each precursor peptide is known
because this information is available in the input peak lists of GAPP
(2, 3, 13). Management of elution time also maximizes the
(currently .mgf, .pkl, or mzXML). Fragment ions with a single charge number of transitions that can be performed without overly
are preferred in MS/MS mode; information on charge state is dis- compromising on sensitivity (13).

Molecular & Cellular Proteomics 8.4 701

 TABLE III

702
MRMaid performance versus experimentally verified transitions taken from Anderson and Hunter (10)
The results shown are derived from MRMaid searches using the default search parameters, namely no internal cleavage sites, 80% of the precursor mass, and 50% of total
observations. Peptides that did not meet the criterion of 50% of total observations (shown as red rows in the table of results) were still considered because of the current quantity
of available MS/MS reference data. An “observation” in this case is an identification made using a single MS/MS data set that was submitted to GAPP, the results of which have been
mined by the MRMaid program. Results are derived from searches performed on August 26, 2008; therefore, the results shown here may have changed because of the increasing
volume of data in the GAPP repository. Values are rounded to one decimal place in the table, but MRMaid shows up to three decimal places in the results on the website.
No. times
No. of peptide Precursor peptide No. of No. of RT MS/MS transition (m/z) Mean relative
peptide Average No. of times
Protein target (Ensembl candidates sequence used by predicted predicted intensity for
seen in TS for product ion
gene identifier) found for Anderson and Hunter b-ions, y-ions product ion
GAPP for peptide Real Predicted Real Predicteda seen
target (10) (1⫹ only) (1⫹ only) observationsb
target
min
Afamin (ENSG00000079557) 9 DADPDTFFAK 9 43.4 5 5 21.5 19.9 825.4 (y7) 825.3 7 100
940.4 (y8) 940.2–940.4 7 35.1
␣1-Acid glycoprotein 1 6 NWGLSVYADKPETTK 14 36.6 5 6 19.7 20.7 1052.5 (y9) 1052.3–1052.5 13 91.5
(ENSG00000187681) 1068.5 1068.6 1 7
␣2-Antiplasmin 5 LGNQEPGGQTALK 13 33.7 7 6 12.6 14.9 771.4 (y8) 771.2–771.4 13 100
(ENSG00000167711)
␣1-Antitrypsin 18 DTEEEDFHVDQVTTVK 55 29.9 9 8 17.4 21 790.4 (y7) 789.9–791.4 53 42.3
(ENSG00000197249) 889.5 (y8) 888.7–890.0 53 96.8

Molecular & Cellular Proteomics 8.4

Designing MRM Transitions Using MRMaid

␣2-Macroglobin 47 LLIYAVLPTGDVIGDSAK 22 29.1 11 8 36.5 28.7 1059.5 (y11) 1059.1–1059.6 21 99.3

(ENSG00000175899) 1172.6 (y12) 1172.2–1172.7 20 25.9
Angiotensinogen 12 ALQDQLVLVAAK 13 36.5 7 5 23.5 24.1 956.6 (y9) 956.3–956.6 13 45.6
(ENSG00000135744) 713.5 (y7) 713.2–713.5 12 20.0
Antithrombin-III 15 DDLYVSDAFHK 10 39.0 5 6 19.2 22.3 803.4 (y7) 803.2–803.4 10 95.1
(ENSG00000117601) 704.3 (y6) 704.1–704.4 10 91.9
Apolipoprotein A-II 3 SPELQAEAK 1 37.3 5 4 12.1 15 546.4 (y5) 546.3 1 12
(ENSG00000158874) 659.4 (y6) 659.4 1 9
Apolipoprotein E 10 LGPLVEQGR 7 48.4 2 5 15.5 17.4 701.4 (y6) 701.2–701.5 7 3.7
(ENSG00000130203) 588.3 (y5) 588.2–588.4 7 21.3
␤2-Glycoprotein 1 10 ATVVYQGER 9 36.0 4 4 12.4 15.4 652.3 (y5) 652.2–652.3 8 100
(ENSG00000091583)
Ceruloplasmin 24 EYTDASFTNR 10 40.2 5 5 14.9 16.5 624.3 (y5) 624.2–624.3 10 93
(ENSG00000047457) 695.3 (y6) 695.2–695.4 10 30.8
Clusterin 13 LFDSDPITVTVPVEVSR 37 29.8 10 9 28.5 26.3 1296.7 (y12) 1296.3–1297.5 36 27.9
(ENSG00000120885)
Coagulation factor XIIa heavy 8 VVGGLVALR 9 44.7 3 6 19.7 21.6 784.5 (y8) 784.4–784.6 7 2.1
chain (ENSG00000131187) 685.4 (y7) 685.3–685.4 8 100
Complement C9 10 AIEDYINEFSVR 13 37.9 5 6 28.3 31.3 1271.6 (y10) 1271.2–1272.2 12 27.1
(ENSG00000113600) 1027.5 (y8) 1027.2–1028.2 13 23
Complement factor H 33 SPDVINGSPISQK 23 34.4 7 6 16.3 17.7 830.4 (y8) 830.1–831.2 17 93.6
(ENSG00000000971) 572.3 (y5) 572.2–572.5 23 30.3
Fibrinogen ␣ chain 16 GSESGIFTNTK 15 40.3 4 7 14.7 17.3 610.3 (y5) 610.1–610.4 13 91.2
(ENSG00000171560) 780.4 (y7) 780.2–781.5 15 31.1
867.5 (y8) 867.2–867.5 13 27.1
Fibrinogen ␤ chain 14 QGFGNVATNTDGK 4 46.7 6 6 13.5 14.6 706.3 (y7) 706.2–706.5 4 89.3
(ENSG00000171564) 805.4 (y8) 805.3–805.4 3 20.3
Fibronectin 32 VTWAPPPSIDLTNFLVR 8 28.13 4 6 38.2 30.4 977.5 (y8) 977.54 1 4
(ENSG00000115414) 862.5 (y7) 862.9–863.2 3 5.3
Haptoglobin ␤ chain 7 VGYVSGWGR 28 41 5 5 18.2 17.9 562.3 (y5) 562.1–562.4 28 100
(ENSG00000197711) 661.3 (y6) 661.1–661.4 28 37.9
Histidine-rich glycoprotein 10 DSPVLIDFFEDTER 29 33.7 7 7 37.7 27.7 1171.5 (y9) 1171.2–1172.25 29 41.6
(ENSG00000113905) 1058.4 (y8) 1058.03 29 94.3
Hemopexin 15 NFPSPVDAAFR 31 59.5 3 6 23.6 20.4 959.6 (y9) 959.2–960.2 14 92.6
(ENSG00000110169) 775.3 (y7) 775.1–775.5 14 8.6
 TABLE III—continued
No. times
No. of peptide Precursor peptide No. of No. of RT MS/MS transition (m/z) Mean relative
peptide Average No. of times
Protein target (Ensembl candidates sequence used by predicted predicted intensity for
seen in TS for product ion
gene identifier) found for Anderson and Hunter b-ions, y-ions product ion
GAPP for peptide Real Predicted Real Predicteda seen
target (10) (1⫹ only) (1⫹ only) observationsb
target
min
Heparin cofactor II 11 TLEAQLTPR 6 52.4 5 5 16.4 18 814.4 (y7) 814.3–814.4 5 100
(ENSG00000099937) 685.4 (y6) 685.3–685.4 5 34.2
Histidine-rich glycoprotein 10 DSPVLIDFFEDTER 29 33.7 7 7 37.7 27.7 1171.5 (y9) 1171.2–1172.2 29 41.6
(ENSG00000113905) 1058.4 (y8) 1058.0–1059.0 29 94.3
Plasma retinol-binding 5 YWGVASFLQK 27 47.5 5 6 35.1 24.1 849.5 (y8) 849.2–849.5 25 100
protein precursor 693.4 (y6) 693.1–693.5 26 49.4
(ENSG00000138207)
Plasminogen 16 LSSPAVITDK 3 51.0 0 6 15.3 17.7 743.4 (y7) 743.3–743.5 3 9.7
(ENSG00000122194) 830.5 (y8) 830.2–830.5 3 14
Prothrombin 16 ETAASLLQAGYK 15 42.4 6 6 20.2 20.6 879.5 (y8) 879.2–879.5 15 55.3
(ENSG00000180210) 679.4 (y6) 679.1–679.4 15 72.8
Serum albumin 30 LVNEVTEFAK 98 49.2 4 6 19.3 19.9 937.4 (y8) 937.1–938.2 95 96.6
(ENSG00000163631) 694.4 (y6) 694.1–694.6 93 21.9
Serum amyloid P-component 7 VGEYSLYIGR 12 53.2 5 6 21.3 20.9 1057.2 (y9) 1057.4–1057.5 10 12.6
(ENSG00000132703) 871.5 (y7) 871.3–871.4 11 49.4
Transferrin 32 EDPQTFYYAVAVVK 37 29.2 8 6 20.3 25.3 1160.6 (y10) 1160.2–1160.9 31 17.6
(ENSG00000091513) 1288.7 (y11) 1288.0–1289.2 17 4.2
Transthyretin 5 AADDTWEPFASGK 78 38.9 7 7 22.3 18.3 921.4 (y8) 921.1–922.2 78 56.7
(ENSG00000118271) 606.4 (y6) 606.1–606.4 77 99.6
Vitamin D-binding protein 15 THLPEVFLSK 14 34.8 4 6 19.7 23.9 819.5 (y7) 819.2–819.5 14 99.4
(ENSG00000145321) 932.5 (y8) 932.2–932.6 14 59.3
Vitronectin 9 DVWGIEGPIDAAFTR 48 32.7 9 8 36.4 42.0 947.5 (y9) 947.1–948.2 42 84.2
(ENSG00000109072) FEDGVLDPDYPR 27 40.6 5 6 22.2 24.9 890.5 (y8) 890.2–891.2 42 33.8
875.4 (y7) 875.1–875.5 26 100
1031.5 (y9) 1031.1–1031.8 26 7.8
Zinc ␣2-glycoprotein 8 EIPAWVPFDPAAQITK 13 29.9 8 8 36.2 27.3 1087.7 (y10) 1087.2–1087.6 12 97.5
(ENSG00000160862) 728.4 (y7) 728.2–728.6 13 7.8
a
These values may be shown as a range to account for the mass tolerance of fragment ions entered by the user when the MS/MS data were submitted to GAPP for analysis.
b
Relative intensity refers to the fact that signal intensity is normalized by the GAPP analysis pipeline. When spectra are uploaded for protein identification, the y-dimension of each
spectrum is normalized to 100. This means that for all the spectra in the GAPP database that are mined by the MRMaid program there is a maximum y value of 100. This is also
reflected in the graphical MS/MS spectra displayed on the MRMaid product ion results page. The intensity values differ for each individual peak in each experiment submitted to GAPP;
therefore the mean over all observations is given for each particular ion species, for example, y8. It is possible to successfully monitor product ions at low abundance as long as there
is no overlap with other peaks: generally there is less likely to be overlap at the very high m/z end of the spectrum; therefore the higher the m/z, the lower the abundance that can
be successfully tolerated.

Molecular & Cellular Proteomics 8.4

Designing MRM Transitions Using MRMaid

703
Designing MRM Transitions Using MRMaid

Implementation of RT was necessary for MRMaid to be- cate their reproducibility and reliability. At this stage users
come a MRM, rather than a SRM, design tool. By having RT may select several of the best product ions. In this way,
indicators to compare across all candidate transitions, the MRMaid supports multiple product ion selection for each
order of transitions in a multiplexed experiment can be peptide. This allows monitoring of a target even in very com-
planned. The problem in achieving this is that GAPP is de- plex samples because having several ions to act as signposts
pendent on data sources in the public domain, and RT infor- for the protein in MS/MS can confirm its presence despite
mation is rarely captured in publicly available data sets. In the high levels of noise. Moreover by allowing users to apply a
absence of these data, we chose to predict RT using a linear sample type filter at the time of search, such as serum,
peptide RT algorithm, which holds true for peptides up to ⬃20 transition design for complex samples is further supported
residues (28). This was acceptable because MRM peptides because this strategy increases the likelihood that a candi-
are typically no more than 24 amino acids. The procedure date will be selected that has been proven to work even
involves summation of residue coefficients and then correc- among associated sample noise.
tion for the length of the peptide and factors relating to the MRM Test Cases—To demonstrate the ability of MRMaid to
procedure and setup of the column. There are eight different predict MRM transitions, a selection of transitions were taken
options for reverse phase column setup provided by MRMaid from the literature and compared with the results generated
for this process, each of which can be selected by the user by the MRMaid program (Table III). Diverse data sets from
using a drop-down menu. There is a default option available, Human Proteinpedia (29), PeptideAtlas, and Tranche were
namely a microflow method (4 ␮l/min) applying a gradient of analyzed and stored in the GAPP database, including serum-
1– 80% ACN over 60 min (1.32% ACN/min) using a 150-␮m ⫻ based identifications required for the test cases.
150-mm Vydac姞 218TP C18 bead column (5 ␮m). We recom-
mend selection of the specific setup that reflects the user’s DISCUSSION AND FUTURE DIRECTION
own RP chromatography procedure. Therefore, there is a MRMaid is a tool for designing MRM transitions and is
warning printed on the results page when the default option is intended to support the proteomics community by exploiting
selected to remind users that they should choose a specific public data resources and prior knowledge of the MRM tech-
setup wherever possible. nique. MRMaid eliminates the need for time-consuming pre-
RT prediction is programmed in a modular fashion so that liminary studies by delivering ranked candidate transitions
as improved RT models become available new algorithms based on an existing repository of experimental data, mean-
may be integrated into MRMaid. No current RT prediction is ing that far fewer transitions need to be validated before a
believed to be truly universal, but despite these limitations, we suitable candidate is found. MRMaid is freely available over
found that the RT and hydrophobicity estimates were fre- the Internet as an executable web-based service.
quently sufficient to avoid co-elution when planning a MRM As with any automated work flow, no human judgment
experiment using MRMaid (Table III). could be applied to interpret the results; only widely applica-
MRMaid for MRM—MRMaid is essentially a SRM design ble rules could be used. Despite this, the MRMaid results
tool that can be used for MRM design by combining the presented demonstrate that accurate peptide-product ion
results of several SRMs in a single spreadsheet. As explained transition predictions can be made when MS/MS data are
above, a protein accession number is entered via the inter- available for querying. Estimation of RT is also of high enough
face, and candidates are predicted. The results, which may be accuracy to be useful for ordering transitions and avoiding
downloaded, include both product ion and precursor data; co-elution.
then following this, the next protein to be targeted is entered. As data submissions to GAPP increase over time, the
Candidates are generated by MRMaid and downloaded as scope for the use of MRMaid will also increase. In conclu-
before. This process is repeated until all the spreadsheet data sion, MRMaid represents an effective first step toward the
have been captured. Finally using transition score, number of future of integrated software applications for the design of
observations, and retention time, the candidates may be or- quantitative proteomics experiments.
dered for experimental validation in MS/MS before purchas-
Acknowledgment—We thank Eleonora Grosso for providing a sub-
ing synthetic surrogates or designing an expression con-
set of the functions used in MRMaid.
struct. Recommendations and tips to assist users in optimal
candidate selection are provided in the interactive help doc- * This work was supported by Biotechnology and Biological Sci-
umentation on the website. ences Research Council, Engineering and Physical Sciences Re-
MRMaid for Multiple Transitions and Complex Sample Tran- search Council, and GlaxoSmithKline plc.
储 To whom correspondence should be addressed. Tel.: 44-1234-
sition Design—MRMaid is particularly suitable for designing
758512; E-mail: [email protected].
multiple transitions. A short list of multiple peptide candidates
is provided for each target so two or three of these can be REFERENCES
selected for validation. For each of these, a list of many 1. Gerber, S. A., Rush, J., Stemman, O., Kirschner, M. W., and Gygi, S. P.
different product ions is provided, each with metrics to indi- (2003) Absolute quantification of proteins and phosphoproteins from cell

704 Molecular & Cellular Proteomics 8.4

 Designing MRM Transitions Using MRMaid

lysates by tandem MS. Proc. Natl. Acad. Sci. U. S. A. 100, 6940 – 6945 16. Lange, V., Malmstrom, J., Didion, J., King, N. L., Johansson, B. P.,
2. Kay, R. G., Gregory, B., Grace, P. B., and Pleasance, S. (2007) The appli- Schaefer, J., Rameseder, J., Wong, C.-H., Deutsch, E. W., Brusniak,
cation of ultra-performance liquid chromatography/tandem mass spec- M.-Y., Buehlmann, P., Bjoerck, L., Domon, B., and Aebersold, R. (2008)
trometry to the detection and quantitation of apolipoproteins in human Targeted quantitative analysis of Streptococcus pyogenes virulence fac-
serum. Rapid Commun. Mass Spectrom. 21, 2585–2593 tors by multiple reaction monitoring. Mol. Cell. Proteomics 7, 1489 –1500
3. Keshishian, H., Addona, T., Burgess, M., Kuhn, E., and Carr, S. A. (2007) 17. Deutsch, E. W., Lam, H., and Aebersold, R. (2008) PeptideAtlas: a resource
Quantitative, multiplexed assays for low abundance proteins in plasma for target selection for emerging targeted proteomics workflows. EMBO
by targeted mass spectrometry and stable isotope dilution. Mol. Cell. Rep. 9, 429 – 434
Proteomics 6, 2212–2229 18. Craig, R., Cortens, J. P., and Beavis, R. C. (2005) The use of proteotypic
4. Zhang, F., Bartels, M. J., and Stott, W. T. (2004) Quantitation of human peptide libraries for protein identification. Rapid Commun. Mass Spec-
glutathione S-transferases in complex matrices by liquid chromatogra- trom. 19, 1844 –1850
phy/tandem mass spectrometry with signature peptides. Rapid Com- 19. Shadforth, I., Xu, W., Crowther, D., and Bessant, C. (2006) GAPP: a fully
mun. Mass Spectrom. 18, 491– 498 automated software for the confident identification of human peptides
5. Kuhn, E., Wu, J., Karl, J., Liao, H., Zolg, W., and Guild, B. (2004) Quantifi- from tandem mass spectra. J. Proteome Res. 5, 2849 –2852
cation of C-reactive protein in the serum of patients with rheumatoid 20. Craig, R., and Beavis, R. C., (2003) A method for reducing the time required
arthritis using multiple reaction monitoring mass spectrometry and 13C- to match protein sequences with tandem mass spectra. Rapid Commun.
labeled peptide standards. Proteomics 4, 1175–1186 Mass Spectrom. 17, 2310 –2316
6. Beynon, R. J., Doherty, M. K., Pratt, J. M., and Gaskell, S. J. (2005) 21. Shadforth, I., Dunkley, T., Lilley, K., Crowther, D., and Bessant, C. (2005)
Multiplexed absolute quantification in proteomics using artificial QCAT Confident protein identification using the average peptide score method
proteins of concatenated signature peptides. Nat. Methods 2, 587–589 coupled with search-specific, ab initio thresholds. Rapid Commun. Mass
7. Unwin, R. D., Griffiths, J. R., Leverentz, M. K., Grallert, A., Hagan, I. M., and Spectrom. 19, 3363–3368
Whetton, A. D. (2005) Multiple reaction monitoring to identify sites of 22. Mallick, P., Schirle, M., Chen, S. S., Flory, M. R., Lee, H., Martin, D., Ranish,
protein phosphorylation with high sensitivity. Mol. Cell. Proteomics 4, J., Raught, B., Schmitt, R., Werner, T., Kuster, B., and Aebersold, R.
1134 –1144 (2007) Computational prediction of proteotypic peptides for quantitative
8. Cox, D. M., Zhong, F., Du, M., Duchoslav, E., Sakuma, T., and McDermott, proteomics. Nat. Biotechnol. 25, 125–131
J. C. (2005) Multiple reaction monitoring as a method for identifying 23. Côté, R. G., Jones, P., Martens, L., Kerrien, S., Reisinger, F., Lin, Q.,
protein posttranslational modifications. J. Biomol. Tech. 16, 83–90 Leinonen, R., Apweiler, R., and Hermjakob, H. (2007) The Protein Iden-
9. Ciccimaro, E., Hevko, J., and Blair, I. A. (2006) Analysis of phosphorylation tifier Cross-Referencing (PICR) service: reconciling protein identifiers
sites on focal adhesion kinase using nanospray liquid chromatography/ across multiple source databases. BMC Bioinformatics 18, 401
multiple reaction monitoring mass spectrometry. Rapid Commun. Mass 24. Yen, C. Y., Russell, S., Mendoza, A. M., Meyer-Arendt, K., Sun, S., Cios,
Spectrom. 20, 3681–3692 K. J., Ahn, N. G., and Resing, K. A. (2006) Improving sensitivity in shotgun
10. Anderson, L., and Hunter, C. L. (2006) Quantitative mass spectrometric proteomics using a peptide-centric database with reduced complexity:
multiple reaction monitoring assays for major plasma proteins. Mol. Cell. protease cleavage and SCX elution rules from data mining of MS/MS
Proteomics 5, 573–588 spectra. Anal. Chem. 78, 1071–1084
11. Rifai, N., Gillette, M. A., and Carr, S. A. (2006) Protein biomarker discovery 25. Falth, M., Savitski, M. M., Nielsen, M. L., Kjeldsen, F., Andren, P. E., and
and validation: the long and uncertain path to clinical utility. Nat. Bio- Zubarev, R. A. (2007) SwedCAD, a database of annotated high-mass
technol. 24, 971–983 accuracy MS/MS spectra of tryptic peptides. J. Proteome Res. 6,
12. Wolf-Yadlin, A., Hautanleml, S., Lauffenburger, D. A., and White, F. M. 4063– 4067
(2007) Multiple reaction monitoring for robust quantitative proteomic 26. Zhang, J., Gao, W., Cai, J., He, S., Zeng, R., and Chen, R. (2005) Predicting
analysis of cellular signaling networks. Proc. Natl. Acad. Sci. U. S. A. molecular formulas of fragment ions with isotope patterns in tandem
104, 5860 –5865 mass spectra. IEEE/ACM Trans. Comput. Biol. Bioinform. 2, 217–230
13. Stahl-Zeng, J., Lange, V., Ossola, R., Eckhardt, K., Krek, W., Aebersold, R., 27. Harrison, A. G., Young, A. B., Bleiholder, C., Suhai, S., and Paizs, B. (2006)
and Domon, B. (2007) High sensitivity detection of plasma proteins by Scrambling of sequence information in collision-induced dissociation of
multiple reaction monitoring of N-glycosites. Mol. Cell. Proteomics 6, peptides. J. Am. Soc. Mass Spectrom. 128, 10364 –10365
1809 –1817 28. Krokhin, O. V., Craig, R., Spicer, V., Ens, W., Standing, K. G., Beavis, R. C.,
14. Lenz, C., Kühn-Hölsken, E., and Urlaub, H. (2007) Detection of Protein-RNA and Wilkins, J. A. (2004) An improved model for prediction of retention
crosslinks by nano LC-ESI-MS/MS using precursor ion scanning and times of tryptic peptides in ion pair reversed-phase HPLC: its application
multiple reaction monitoring (MRM) experiments. Am. Soc. Mass Spec- to protein peptide mapping by off-line HPLC-MALDI MS. Mol. Cell.
trom. 18, 869 – 881 Proteomics 3, 908 –919
15. Barnidge, D. R., Dratz, E. A., Martin, T., Bonilla, L. E., Moran, L. B., and 29. Mathivanan, S., Ahmed, M., Ahn, N. G., Alexandre, H., Amanchy, R.,
Lindall, A. (2003) Absolute quantification of the G protein-coupled recep- Andrews, P. C., Bader, J. S., Balgley, B. M., Bantscheff, M., Bennett,
tor rhodopsin by LC/MS/MS using proteolysis product peptides and K. L., Björling, E., Blagoev, B., and Pandey, A. (2008) Human Proteinpe-
synthetic peptide standards. Anal. Chem. 75, 445– 451 dia enables sharing of human protein data. Nat. Biotechnol. 26, 164 –165

Molecular & Cellular Proteomics 8.4 705