Proteinmatching Analysis Software™ (PAS™)

PAS™ is part of the MBS™ bioinformatics package!

GS4FTMS, DeNoS, PhosTShunter, and ModifiComb were developed by (and PAS™ was developed in collaboration with) Doctor Michael L. Nielsen, Doctor Mikhail M. Savitski and Professor Roman A. Zubarev.

In the growing field of proteomics identification of proteins by tandem mass spectrometry (MS/MS) is performed by matching experimental mass spectra against calculated spectra of all possible peptides in a protein database.

The search engine assigns to each spectrum a "score" which indicates how well the experimental data complies with the expected data; higher score means increased confidence in the identification.

One problem with this approach is the false positive identification, which arise from incomplete data, as well as from the presence of misleading ions in experimental mass spectra due to gas-phase reactions, stray ions, contaminants and electronic noise. MS-based proteomic experiments are further affected by a rather poor efficiency typical in the range of 10-15%, implicating that only a low percentage of acquired mass spectrometric data is significantly identified and assigned a peptide sequence. We employed a novel technique for reduction of false positives, which is based on a combined use of orthogonal fragmentation techniques electron capture dissociation (ECD) and collisionally activated dissociation (CAD). This approach not only shows a significant improvement in protein identification confidence, but also additionally reduces the presence of false-positive identifications while increasing the overall efficiency rate of proteomics experiments.

For reliable identification of even known proteins, complete de novo sequencing of their peptides is desired. The main problems of conventional sequencing based upon MS/MS are incomplete backbone fragmentation and the frequent overlap of fragment masses. The first proteomics-grade de novo approach has therefore been developed, where the above problems are alleviated by the use of complementary fragmentation techniques CAD and ECD. An approach that combines hardware and software improvements to achieve a validity level corresponding to >95% correct de novo assignments of peptide sequences.

Post-Translational Modifications (PTMs) are key regulators of protein function, localization, and interactions taking place inside the cell. Post-translational modification is the chemical modification of a protein after its translation. It is one of the later steps in protein biosynthesis for many proteins. PTMs are also required for proper folding of the protein. The extent of modifications is very important for shotgun proteomics that is based on identification of peptides derived from enzymatically digested complex protein mixtures. The shotgun approach is known to face the so-called dynamic range challenge arising from the fact that concentrations of proteins in whole proteomes or complex mixtures such as blood plasma differ by many orders of magnitude. The challenge therefore is to detect low-abundance peptides in the presence of much more abundant competitors, but extensive modifications can exacerbate this problem significantly. As traditional database search approaches allow for only few modifications to be included in the search, protein identification in shotgun proteomics is largely based on the detection of unmodified polypeptides.

Identification of all PTMs present in a sample is highly relevant from a biological point of view. Traditionally PTM studies have been following the idea that one modification site corresponds to one regulatory function, and have therefore focused on specific amino acid residues or a small number of residues in a specific protein. However, instead of single-site modifications being responsible for protein activity and stability, it is now evident that proteins can be modified at several sites functioning together – a phenomenon referred to as multisite modifications Full sample analysis with identification of all acquired MS/MS spectra is therefore necessary in order to characterize the full extent and functional importance of all protein modifications in an entire proteome. A reliable method called ModifiComb for unbiased mapping of hundreds types of PTMs at a time, including novel and unexpected PTMs was developed in our laboratory to achieve this goal.

Publications concerning GS4FTMS, DeNoS, PhosTShunter and ModifiComb

GS4FTMS

GS4FTMS (golden and complementary pair extraction from collisionally activated dissociation CAD and electron capture dissociation ECD data).

Task:

GS4FTMS extracts consensus data from CAD and ECD mass spectra. GS4FTMS decides the "direction" of the peptide fragment (y or b ions). The software makes use of the high mass accuracy to decide what a peptide fragment is and what is not. It also decides the charge of all peptide fragments and transforms all m/z values into neutral masses.

Advantage:

GS4FTMS performs faster database searches. GS4FTMS also performs more reliable peptide identifications and identifies 50% more proteins.

Input:

DTA files, where each file contains data from a mass spectrum, either ECD or CAD.

Output:

For each pair of ECD and CAD files an output file is generated containing consensus data.

DeNoS

DeNoS (de novo sequencing of peptides fragmented using CAD and ECD).

Task:

DeNoS performs complete or almost complete sequencing of peptides with reliability (>95%). DeNoS uses all information from CAD and ECD spectra. It is a hierarchal algorithm. In the first step fragments that are confirmed in both CAD and ECD (so called Golden Complementary Pairs) along with fragments that are only found in CAD (so called Complementary Pairs) are used. After that, step-by-step fragments with low reliability are used. In the last step, if the peptide is still not fully sequenced, the software uses a trivial application from the graf theory to sequence the remaining peptide parts with "unreliable" fragments.

Advantage:

DeNoS is the first algorithm ever to be able to sequence peptides with >95% reliability. 13% percent of all MS/MS spectra are almost completely sequenced (in typical experiments you usually only identify about 10% of all MS/MS spectra using a search engine, so 13% in this case is very good).

Input:

DTA files, where each file contains data from a mass spectrum, either ECD or CAD.

Output:

Complete or almost complete peptide sequences.

PhosTShunter

PhosTShunter (detects all phosphorylated peptides in MS/MS data).

Task:

The software uses 10 different requirements to test if a peptide is phosphorylated or not. The combination of two of these requirements is in general able to find all phosphorylated peptides, without giving any false suggestions. The remaining eight requirements are less specific but they give a good indication about the quality of the spectra.

Advantage:

PhosTShunter detects in general 2 times as many phosphorylated peptides as Mascot. The software is also much faster than Mascot.

Input:

DTA files, where each file contains data from a mass spectrum, either ECD or CAD. The software can also work with just CAD files.

Output:

A spectrum number and a vector of the length 10 with information about how the 10 different requirements have been fulfilled.

ModifiComb

ModifiComb (detects all types of peptide modifications, for known and unknown peptides).

Task:

ModifiComb compares spectra of identified peptides with spectra of unidentified peptides. If there is a significant resemblance between these spectra then the software will report that the unidentified peptide is a modified version of the identified peptide along with a modification mass that is the difference between the two peptide's molecular mass. A flow-schedule of the software is shown in the picture below.

Advantage:

ModifiComb is a unique method with high reliability (>97%) and high sensitivity. ModifiComb can identify all types of modifications in a proteomics experiment. The software does not only identify all peptide modifications but it's also much faster than other search engines that are only able to detect known peptide modifications.

Input:

DTA files, where each file contains data from a mass spectrum, either ECD or CAD. The software is also able to work with just CAD files. The software also requires a list of sequences and spectra numbers of identified peptides. However, a certain version of the software is able to work without the list of identified peptides, so called "blind search".

Output:

The outcome contains sequences of modified peptides, masses of modifications, and proposals for where the modifications can be found. High resolution data from whichever histogram of the sample's modification profile can also be generated.

ModifiCombCAD

ModifiCombCAD (detects all types of peptide modification, for known and unknown peptides).

Task:

ModifiCombCAD works in the same way as ModifiComb, but expects its own data.

Modules & Sub-Modules

PAS™ contains four (4) main Modules. These Modules are named GS4FTMS, DeNoS, PhosTShunter and ModifiComb.

Module 1: GS4FTMS

GS4FTMS performs faster database searches. GS4FTMS also performs more reliable peptide identifications and identifies 50% more proteins.

Module 2: DeNoS

DeNoS is the first algorithm ever to be able to sequence peptides with >95% reliability. 13% percent of all MS/MS spectra are almost completely sequenced (in typical experiments you usually only identify about 10% of all MS/MS spectra using a search engine, so 13% in this case is very good).

Module 3: PhosTShunter

PhosTShunter detects in general 2 times as many phosphorylated peptides as Mascot. The software is also much faster than Mascot.

Module 4: ModifiComb

ModifiComb is a unique method with high reliability (>97%) and high sensitivity. ModifiComb can identify all types of modification in a proteomics experiment. The software does not only identify all peptide modification but it's also much faster than other search engines that are only able to detect known peptide modifications.