Mass spectrometry gained a wide popularity among biologists and biochemists because all three main kinds of biopolymers - proteins, nucleic acids and saccharides - can be efficiently desorbed/desolvated and ionized for analysis in a variety of analyzers. Automated computer-based solutions were devised for all three biopolymers, leading in particular to high throughput proteomics and to glycomics. These computer systems are usually shipped by mass spectrometer vendors along with their instruments.

However, numerous fields of endeavor cannot benefit from such automation frameworks: specially-crafted software is needed. This is particularly true for studies involving the fine structural details of biopolymeric analytes. Studies involving protein post-translational modifications or structure/function relationships essentially are not automatable, requiring human inspection of each and every mass spectrum so as to make the most out of the data.
The massXpert software program was developed in the late nineties to fill a gap in the software offerings for protein mass spectrometry (massXpert 1, [1]). In particular, a software program was needed which would allow 1. to define any number of post-translational modifications and 2. to easily apply any given modification to any residue in the sequence - with automatic mass recalculation.
Indeed, evaluating structural hypotheses to match mass spectrometric data had to be an easy and quick process. Because biopolymer chemistry (sample preparation techniques, specifically) and mass spectrometry evolved during the last ten years, massXpert had to undergo profound changes to match these technological and methodological advances. In this paper, I will review the major improvements and feature additions that have characterized the development of the latest version of the massXpert software tool, from its initial development ten years ago [1], through one full rewrite (GNU polyXmass, [2]) and a second one (massXpert 2, [3]).

Multiplicity of Polymer Chemistries
Mass spectrometric analysis of saccharides and nucleic acids became mainstream recently, while proteins and peptides had been routinely analyzed for almost two decades.

massXpert 1 only dealt with proteins, with all the chemical logic being hard-wired in the software code. massXpert 2 is the fruit of a full redesign of the polymer chemistry handling, with a new layer interfacing the mass calculation engine and the sequence editor where simulations are performed. The user is provided with an interface where to model the polymer chemistry by defining the chemical entities needed, from atoms to cleavage and gas-phase fragmentation specifications through monomers and modifications (fig. 2).

Electrospray Ionization-based Simulations: Multi-charged Ions
MALDI-TOF was the main source-analyzer combination in use at the time massXpert 1was developed, and that software version only handled monoprotonated proteins and peptides. Recent developments have pushed electrospray ionization forward, with the resulting production of abundant multicharged ion data. massXpert 1 was virtually useless in an electrospray ionization setting, and thus multiple-charge ionization simulations were implemented throughout the new massXpert 2 version. Equally useful is the ability of massXpert 2 to easily switch ionization agent, from protonation to cationization with sodium, for example, which is of utmost interest when mining saccharide or nucleic acid mass data.

New Cross-link Framework
New ways to perform mass spectrometric studies of intra-molecular crosslinks have been possible since the use of electron capture/transfer dissociation (ECD/ETD), which involves the rupture of amide bonds while preserving post-translational modifications. MassXpert 2 now incorporates an innovative cross-link framework where it is possible to define any number of cross-links to be set between any number of monomers in a polymer sequence. The mass calculation engine can be configured to take these cross-links into account or not (fig. 1). Sophisticated cross-links may be defined, such as those involved in the formation of the chromophore in fluorescent proteins [4]. Further, the cross-links are taken into account when performing polymer sequence cleavages, including when partial cleavages are required: this is no small feat from an algorithmic standpoint.

Monomer Chemical Modifications Are Better Simulated
While protein post-translational modifications were one of the reasons massXpert was developed in the first place, a number of discoveries since ten years prompted a redesign of the monomer modification framework. In particular, it was decided that any monomer should be able to hold any number and type of modification. This has become an essential feature to model histone chemistry, for example, with lysine residues that can undergo up to three methylations (fig. 3).
Provisions can be set in the polymer chemistry definition to ensure that a polymer sequence does not get modified in an irrealistic manner.