Description of the intermediary file for import of MD trajectories into CSDB
document ver. 1.15

In parallel .json files with the same name are generated upon trajectory processing, and store data for visualisation: abundance map with meta-data, energy map with meta-data, torsion sequences in trajectory
The root tag in XML is <md_data>..</md_data>; all other data reside inside.

TRAJECTORY PARAMETERS (XML) with example values

<confmap>..</confmap>
# abundance conformation map (1D, 2D, or 3D)
# phi is tab-separated across, psi is newline-separated along, omega are blocks separated by a blank line

<lincode>
aDGlcp(1-2)[Me(1-4)]aDGlcp
</lincode>
# saccharide structure in CSDB Linear notation (normalized 'lincode')

<method>MM3-2000</method>
# method/force field used in MD simulation (MM3-2000, GLYCAM, CHARMM36, MMFF94, OPLS-AA etc.)

<temperature>300</temperature>
# temperature of MD simulation, in Kelvins

<extremum_step>
Phi: 5
Psi: 5
Omega:15
</extremum_step>
# extremum search dihedral step value, degrees (a positive divisor of 360 - 2,3,5,10,15,20,30, etc.)
# (= angular precision of torsions) (=из карты с каким шагом получены минимумы)

<confmap_step>
Phi: 10
Psi: 10
Omega:30
</confmap_step>
# map dihedral step value, degrees (a positive divisor of 360 - 2,3,5,10,15,20,30, etc.)
# (= angular precision of torsions)

<start_bound>..</start_bound>
#start bound of a conformation map, degrees. Default = -120

<end_bound>..</end_bound>
#end bound of a conformation map, degrees. Default = 240

<frames>100000</frames>
# number of frames in the trajectory file

<solvent>NONE</solvent>
# solvent model (NONE, GB, Stil, TIP3P, explicit) (=для Glycomaps всегда NONE)

<extrema>..<extrema>

#inter-residue linkage conformations, one line per energy minimum
#the tab-separated fields are linkages sorted according to the rules below, abundance, and relative energy
#abundance is a positive integer  indicating the population of the conformation
#energy is its energy in kcal/mol, referenced to the global minimum as 0. Usually, energy is within 0.0..5.0 kcal/mol
#each linkage is either empty (if no torsions has been derived from MD trajectory) or a comma-separated list. This list can have one (phi), two (phi,psi), three (phi,psi,omega) or four (phi,psi,omega,theta) angles according to the number of ratatable bonds in the linkage. The angles are usually integers in the range -120..239
#the order of angles within a linkage is from donor residue to acceptor residue, e.g. in aDManp(1-3)aDGlcp phi is at mannose side, and psi is at glucose side. If donor/acceptor cannot be distinguished (e.g. in sucrose), donor is a residue at the left in CSDB Linear encoding.
#more details about naming of angles and involved atoms are available in the publication [REF].
#the angle N acually means a range of angles N..N+step

#EXAMPLES:

#aDGlcp(1-2)[Me(1-4)]aDGlcp 
#acording to the sorting rules (see below), two-torsion interglycosidic linkage is linkage 1, and one-torsion methyl linkage is linkage 2.
#there are three minima in the conformation space, the first of them does not have data on methyl linkage, while the two latter ones have.
#linkage 1	linkage 2	abundance	energy
180,0		50	-3.6 #это пример, когда вторая связь не посчитана (напр., связь с ацетатом в структурах из Glycomaps)
180,10	10	80	-4.2
180,20	10	200	-4.9

#aDGlcp(1-6)bDManp
#in this example, there is only one three-torsion linkage, and conformation space has two minima (phi,psi,omega)
#linkage 1	abundance	energy
-55,25,60	1500	-4.5
-155,35,60	500	-4.0

#aDGlcp(1-3)[Ac(1-2)]aDGalpN
#acording to the sorting rules (see below) acetate linkage is linkage 1, and two-torsion interglycosidic linkage is linkage 2
#in this example conformation space has three minima (phi,psi),
#and, linkage with Ac is not calculated (e.g. as in the processed trajectory from Glycomaps), so its has empty fields in column 1
#linkage 1	linkage 2	abundance	energy
	55,20	50	-3.6 
	65,30	80	-4.2
	220,-25	200	-4.9


# LINKAGE SORTING (affects the order of columns above)

# this rules are exmplified on a sample structure C(1-2)[E(1-4)]B(1-2)[D(1-4)]A, where capital letters are residue.
# in a tree graph donors are children, acceptors are parents, root is a node without parents (residue A).
# all linkages point downtree; a node can have zero or more incoming linkages (from its donors), and it can have zero (if root) or one outgoing linkage (to its acceptor)
# linkages are ascendygly-sorted by residue, then by substitution position
# incoming linkages of the same residue are sorted by substitution position (atom number), e.g. for residue A first linkage is B(1-2)A, the second linkage is D(1-4)A
# the residues are sorted the same way as linkages, i.e. residues attached to lower positions go first. In the example structure residue order becomes A->B->D->C->E
# So, the resulting sorting for the above structure is:
#linkage 1	linkage 2	linkage 3	linkage 4
#B(1-2)A	D(1-4)A	C(1-2)B	E(1-4)B