CARA vs Xeasy

From Wiki

Share/Save/Bookmark
Jump to: navigation, search

Below is concise list of key features of CARA as compared to XEASY that should help new users make a transition.

Contents

Data Organization

XEASY keeps data stored in separate files: spectra, sequence lists, atom lists, and peaklists.

CARA stores all data in a "repository" - an XML file, which contains all information (residue type definitions, spectrum types definitions, spin systems, spins, peak lists, extension scripts, paths to NMR spectra, etc.). This file typically has a .cara extension. A repository is built from a template by adding "projects". This approach allows much better control and overview over a structure project and makes sharing data among researchers easier.

Though CARA protects repositories from modifications, which may compromise its stability, repositories can be edited with a plain text or XML editor.

Atom Nomenclature

Current repository templates are designed to be compatible with the BMRB format. That is, H is used instead of HN, and glycines have HA2/3 instead of HA1/2. This is also the default format of cyana.lib in CYANA 2.1.

In addition to that, pseudoatoms are labeled as H* instead of Q* and QQ*. For example, HB is used for QB, HD1/2 for QD1/2, and HD for QQD of Leu. This is compatible with the pseudo=2 setting in CYANA 2.1

Residues and Spin Systems

XEASY stores information about spin systems (also referred to as fragments in XEASY manual) and residues in the same *.seq file, and requires somewhat confusing swapping of mapping and fragment numbers for sequence-specific assignment.

CARA clearly distinguishes residues and spin systems as separate classes in repository database. Spin systems can be numbered from 1 upward. They can be linked into fragments by setting successor and predecessor tags. Fragments can be "mapped" to the sequence to see if they match a particular stretch of residues. A spin system becomes assigned when its assignment tag is set, pointing to a certain residue in the sequence.

Atoms and Spins

What XEASY refers to as atoms are called "spins" in CARA. "Atoms" in CARA form a different data class - it is a member of "residue type" as a part of molecular structure. The database of spins can be exported as an XEASY atom list, and an XEASY atom list can be imported into CARA (provided that the nomenclature matches).

While you may need to use separate atom lists with different spectra and peak lists in XEASY, CARA works with a single set of spins. Spin aliases are used in situations where the same spin has different chemical shifts (see below).

Spin Labels

Spin labels have the form [?|!]SPINLABEL[+|-x], where SPINLABEL is an alphanumeric string and +x or -x is the "offset". The question mark ? as the first character marks temporary assignments, the exclamation sign "!" - stereospecific assignments. Since + and - characters are reserved to specify offsets, lowercase letters "p" and "m" should be used instead for spin labels in GFT spectra. For example, sequential CA+CB peak should be labeled CApCB-1.

Offset designations in CARA carry more information than in XEASY - they are used to search for sequential neighbors.

Peak Inference and Peak Lists

Though peak lists in XEASY format can be exported from any spectrum, they only need to be generated for NOESY spectra to provide input to programs like CYANA and AutoStructure. They are not needed for backbone and side-chain assignment.

Peak marks, which you see in spectrum display tools ( "scopes", in this case other than MonoScope ) are displayed at positions, inferred from chemical shifts, and residue type and spectrum type definitions. Spectrum type definitions describe the correlations prouced in a given spectrum and valid spin labels in every dimension. This is equivalent to XEASY peaklists being generated from an atom list on-the-fly. If a peak is moved in CARA, then the chemical shifts of the spins involved are updated instantly. If a new peak is picked, new spins are created. These changes are synchronized across all "scopes".

Spectral Folding and Aliasing

In CARA all spins and peaks have their exact assignments and positions (in ppm). There are no folding attributes for peaks, as in XEASY. Instead, it is possible to navigate beyond the spectrum edge to display the true unfolded location.

Spin Links

Spin links do not have a direct equivalent in XEASY. They are designed to describe correlations observed in NOESY spectra, but CARA does not prescribed a unique use for them. Spin links present a good means of visualizing UPLs or short distances in NOESY spectra. Spin links can also be used instead of peak list in manual NOE assignment.

A useful feature is that a single spin link produces two peaks - the direct and transposed. Spin links and inferred peaks can be exported as XEASY peak lists.

Spin Aliases

There may be significant mismatches between related NMR spectra. Typical causes are different sample and experimental conditions, such as temperature differences or non-resonant effects, or systematic offsets, such as those between TROSY and conventional NMR spectra. In such cases spin aliases can be helpful.

The spin alias object is a child of spin. It has its own chemical shift (typically different from that of spin itself), and a tag with the ID of the spectrum, where this alias was set. Spin aliases have a purely cosmetic effect on spectral appearance, defining positions, where planes and and slices will be taken.

It is recommended to have all spectra calibrated and matched as best as possible, before resorting to spin aliases. Also you would want to have the main chemical shifts set to match the NOESY spectra, since they are exported for use in automated structure calculation, and use aliases for other spectra, such as HCCH-COSY.

Scripting

CARA functionality is extended by scripts in Lua programming language (www.lua.org). Besides default Lua functionality interface to many CARA functions is available.

-- AlexEletski - 06 Mar 2007

Personal tools