statistics | stat

Calculate various statistics for given data sets.

stat {<name> | ALL} [shift <value>] [out <filename>] [noeout <filename>] [ignorenv] [name <noe setname>]

<name> Name of data set to analyze. ALL analyze all data sets.
shift <value> Subtract <value> from all elements in each data set.
[out <filename>] Write analysis results to <filename> (STDOUT if not specified).
[noeout <filename>] (Type ’noe’ only) Write summary of NOE results to <filename>.
[ignorenv] (Type ’noe’ only) Ignore negative NOE violations (i.e. shorter-than-expected distances).
[name <noe setname>] (Type ’noe’ only) Name for output NOE data sets.

DataSet Aspects for type ’noe’ output:
[R6] Averaged 1/r^6 distance for each set.
[NViolations] Number of violations based on given bounds for each set.
[AvgViolation] 1=r6 averaged distance minus expected distance for each set.
[NOEnames] Name of each set.

Analyze angles, dihedrals, distances, and/or puckers and calculate various properties. More specific analyses can be obtained by labeling distances/dihedrals/puckers (from e.g. the distance, dihedral, pucker commands or with the dataset command) with the type <label> keyword:

dihedral type labels: alpha, beta, gamma, delta, epsilon, zeta, chi, c2p h1p, phi, psi, omega, pchi
distance type labels: noe
pucker type labels: pucker

For each input data set, the average, standard deviation, initial and final values will be reported. The cyclic nature of dihedral/pucker data sets is taken into consideration when averaging.

Torsion Analysis
A table will be written in ASCII format showing the distribution of torsion values for each data set. More specific information may be printed based on the set type. Values in the output marked SNB are from those defined by Schneider, Neidle, and Berman. For more information on nucleic acid torsion as pertains to RNA see further
work by Schneider et al.

For example, to perform in-depth analysis on some nucleic acid dihedral angles:

dihedral g0 out dihedrals.dat :1@O5’ :1@C5’ :1@C4’ :1@C3’ type gamma
dihedral d0 out dihedrals.dat :1@C5’ :1@C4’ :1@C3’ :1@O3’ type delta
dihedral c0 out dihedrals.dat :1@O4’ :1@C1’ :1@N9 :1@C4 type chi
analyze statistics all out stat.dat

Distance Analysis
A table will be written in ASCII format showing the distribution of distance values < 6.5. If a distance is labled as ’type noe’ a compact time series will be printed in ASCII format showing the NOE as strong, medium, or weak. In addition the <r^-6>^(-1/6) averaged value will be reported, as well as the number of upper/lower bound violations. If ’noeout’ is specified, a summary of these results will be written with format:

<#NOE> <R6> <Nviolation> <AvgViolation> <Name>

Where <#NOE> is an index, <R6> is the <r^-6>^(-1/6) averaged distance, <Nviolation> is the total number of bounds violations, <AvgViolation> is the average difference from expected distance Rexp when the distance is violated (note that if not explicitly set, Rexp is set to the upper bound when the lower bound is 0.0, or the average of upper and lower bounds otherwise), and <Name> is the data set legend.

For example, the following input could be used to check certain distances for NOE violations:

distance :3@HB= :10@HG= type noe noe_medium
distance :3@HE= :10@HG= type noe noe_strong
distance :3@HA :12@HA type noe noe_medium
distance :3@HD= :12@HG= type noe noe_medium
distance :3@HE= :12@HA type noe noe_strong
analyze statistics all out dpdp.noe.dat noeout noe_graph.dat name Res3_NOE

Pucker Analysis
A table will be written in ASCII format showing the distribution of pucker phases for each data set.