Documentation applies to versions 11.1.9 and later.
NMRFx Structure can be used to generate and analyze macromolecular structures and predict chemical shifts. It can be run in two different ways. First, if the command is invoked with one of several subcommands (batch, gen, predict, score, summary, super, or train) a predefined mode will be executed. Alternatively, one can simply give as an argument a Python (actually Jython) script which will be executed. The script has access to standard Python functions (including the standard Python library) and to specific commands provided by the NMRFx Structure program's Java code. The predifined subcommands are listed below along with documentation for usage and an example for demonstration.
usage: nmrfxs batch|gen|predict|score|summary|super|train|script.pynmrfxs gen [ -s gen|all|refine ] [ -s seed ] [ -d directory ] [ -r report ] [ projectFile ] [ script.py ]Generate a single structure using data specified in a project file and initializing the random number generator with a specified seed. This is useful for testing out the project file before generating a whole family of structures with the batch command. An output directory will be created if not present. After successful execution, the generated PDB and violation files will be written to that directory. The output files will have the seed number appended to them (i.e. temp0.pdb, temp0.txt, etc.). By default, the output directory gets placed in the current directory, however, the relative path to a different directory can be specified using the directory option. When debugging a structure generated, it may be useful to view the energy violations for all the constraints specified in the project file. To do this, specify the report option. This will output constraint violations at prepartion stage into a file named energyDump$seed_prep.txt within the output directory. Lastly, define torsional angle molecular dynamic procedures in an executable script to replace the builtin annealing protocol. If a python script is specified as the last argument of the command, the script will be executed. The script can alternatively be placed inside the project file replacing the annealing data block.
Generating a structure requires specifying the molecular information, restraints and parameters for the rotational dynamics. This can be done by specifying a YAML file and/or a NEF file. YAML files are human readable data files that hava a relatively simple structure without a lot of "fluff". Project yaml files are described below.
The protocol used for structure calculation proceeds through a seris of stages . These stages involve preparation, high temperature dynamics, simulated annealing and low temperature dynamics. The stages are predefined in code and have specific values for Parameters and Forces. The user can adjust any of the parameters in a stage or introduce new stages by adding sections to the yaml file.
Structure calculations are typically constrained by experimental data. NMRFx supports a variety of formats for importing these restraints. See the Data Section for more details.
Running a structure calculation will generate output as the structure calculation proceeds. This output will be written to standard output and an example is shown here.
Examples:
nmrfxs gen -s 0 project.yaml
nmrfxs gen -s 0 -d ~/gen-structures project.yaml
nmrfxs gen -s 0 -f data.nef
nmrfxs gen -s 0 -f data.nef project.yamlgen OPTIONS:
nmrfxs batch [OPTIONS] projectFileGenerate a family of structures using the data specified in a project file. An output directory and final directory will be created if not present. All generated structures will be written to files (temp1.pdb, temp2.pdb, ...) in the output directory. A violation file (temp1.txt, temp2.text ...) will also be written. The best structures, along with their violation files, will be written to the final directory. Multiple files are generated by repeatedly invoking the nmrfxs gen command with the specified project file and an incremented seed numbers. The number of invocations running simultaneously will be specified by the -p option.
Generating multiple structures with the batch subcommand will generate a variety of output files. These are described here.
batch OPTIONS:
Example:
nmrfxs batch -n 100 -k 10 -p 5 -a project.yaml
NMRFx Structure can predict chemical shifts of proteins and RNA and arbitrary small, organic molecules. Protein predictions are done using geometric attributes (primarily dihedral angles and ring-current shifts). RNA predictions can be done using geometric attributes or secondary structure attribute based methods. Attribute based methods for RNA can be done with a sequence and dot-bracket notation specified in a .yaml file or they can be done by automatic recognition of the secondary structure from the 3D coordinates of the RNA read in from a strucutre (pdb or cif) file. Small molecules are predicted with a HOSE code like method.
predict OPTIONS:
Protein predictions are done for most hydrogen, carbon and nitrogen atoms. RNA geometric predictions are done for all carbon bound protons and their carbons. RNA attribute predictions are done for non-exchangable protons and their parent carbon and nitrogen atoms.
Prediction output in star format looks like the following. This is an NMR-STAR3 chemical shift saveframe without the header.
1 . 1 1 1 64 GLU N N 15 120.210 2.470 . 1 . . . . 64 GLU N . . 1
2 . 1 1 1 64 GLU CA C 13 56.100 0.910 . 1 . . . . 64 GLU CA . . 1
3 . 1 1 1 64 GLU HA H 1 4.130 0.220 . 1 . . . . 64 GLU HA . . 1
4 . 1 1 1 64 GLU CB C 13 29.470 0.950 . 1 . . . . 64 GLU CB . . 1
5 . 1 1 1 64 GLU HB2 H 1 2.000 0.190 . 1 . . . . 64 GLU HB2 . . 1
6 . 1 1 1 64 GLU HB3 H 1 2.070 0.200 . 1 . . . . 64 GLU HB3 . . 1
7 . 1 1 1 64 GLU CG C 13 35.940 0.850 . 1 . . . . 64 GLU CG . . 1Prediction output in protein format looks like the following. This only includes "backbone* atoms, even if additional atoms have predictions:
Num Name N CA CB C H HA(2) HA3
64 GLU 120.21 56.10 29.47 175.45 _ 4.13 _
65 ASP 122.16 54.53 40.80 175.46 8.55 4.74 _
66 PHE 119.86 54.86 39.16 172.19 9.03 4.76 _
67 PRO _ 61.49 30.54 174.44 _ 4.61 _
68 PRO _ 62.14 31.80 175.47 _ 4.95 _RNA predictions based on attributes can be output in star format (as above) or the output can be a list of atom specifiers (residueNumber.atomName), predicted shifts and various attriburtes about the prediction.
Output example:
20.C2' 75.34 N 5 Mean 75.31 +/- 0.24 Range: 74.90 -75.45 Pp_AU_GC_CG_pP_-_-_-_-_-_-
20.H2' 4.43 N 5 Mean 4.43 +/- 0.01 Range: 4.41 -4.44 Pp_AU_GC_CG_pP_-_-_-_-_-_-
20.C1' 92.71 N 6 Mean 92.70 +/- 0.13 Range: 92.55 -92.83 Pp_AU_GC_CG_pP_-_-_-_-_-_-
22.N4 97.53 N 6 Mean 97.57 +/- 0.42 Range: 97.22 -98.38 Pp_CG_CG_-_-_-_-_-_-_-_-
22.H41 8.13 N 12 Mean 8.05 +/- 0.68 Range: 6.84 -8.67 Pp_CG_CG_-_-_-_-_-_-_-_-
22.H42 7.30 N 11 Mean 7.41 +/- 0.64 Range: 6.95 -8.50 Pp_CG_CG_-_-_-_-_-_-_-_-
22.C5 98.24 N 18 Mean 98.24 +/- 0.29 Range: 97.86 -99.20 Pp_CG_CG_-_-_-_-_-_-_-_-
22.H5 5.46 N 34 Mean 5.46 +/- 0.10 Range: 5.23 -5.77 Pp_CG_CG_-_-_-_-_-_-_-_-
22.C6 141.97 N 17 Mean 142.02 +/- 1.10 Range: 141.27 -146.07 Pp_CG_CG_-_-_-_-_-_-_-_-The output for small molecules is a simple table of atom name and chemical shift:
C6 136.60
C7 129.30
C8 157.70
C9 168.20
C10 45.40Examples:
nmrfxs predict protein.pdb
nmrfxs predict -o protein protein.pdb
nmrfxs predict -o star -r attr rna.pdb
nmrfxs predict project.yamlnmrfxs score [OPTIONS] projectFile [pdbFile1.pdb, pdbFile2.pdb, ...]Analyze the quality of the structure(s) generated by using the score subcommand. Note: the summary command listed above analyzes the output files from a previous run of nmrfx batch. This command will load pdb files and analyze them according to the constraints referenced in the .yaml and on the command line (see options below).
score OPTIONS:
Examples:
nmrfxs score -y project.yaml pdb/\*.pdb
nmrfxs score -y project.yaml -p 'pdb/\*.pdb'nmrfxs summary [final/final1.txt, final/final2.txt, ...]summary OPTIONS:
Analyze output files and create a summary file showing what constraints are violated. If no output files are specified as arguments, all final*.txt files in the final subdirectory of the current directory will be analyzed. Summary is different than the score subcommand in that the violation information is harvested from the output files from the structure calculation, whereas the score command loads in pdb files and analyzes them.
The output will be placed in a file named analysis.txt and wil have a format like this:
Type Atom1 Atom2 nViol Bound Mean Max Structures
Dis: 1:141.H3' - 1:141.H8 3 3.00 0.45 0.61 |+++.|
Dis: 1:141.H2 - 1:142.H1' 3 5.00 0.49 0.69 |+++.|
Dis: 1:130.H2 - 1:134.H8 3 5.00 0.25 0.39 |++.+|Calculate the superpostion of a set of models. Multiple cycles (specified with -n flag) of superpostion are done to identify regions that are considered core. Core residues are identified as those residues whose rms value is less than twice the median of the rms for all residues. Superimmposed structures can be output into multiple pdb files or a single MMcif file.
super OPTIONS:
Information pending...
If a python script is provided, instead of one of the above subcommands the script will be executed.
The YAML files contain information about the molecular structure, restraints (distances, angles etc.) and parameters of the dynamics trajectory. It is perhaps best illustrated with a few examples.
The following example indicates that a sequence file should be read from the file 1d3x.seq in the input folder and is in nv format. The nv format is that used by NMRViewJ and is similar to that of CYANA. Angle and distance restraints are read from the specified files, which are in XPLOR format.
The structure calculation happens through a process of simulated annealing with rotational dynamics and only happens if there is an anneal section in the file. The dynOptions section specifies that 15000 steps of dynamics will be done starting at a temperature of 5000. Note the force field etc. is not calibrated such that temperature values have physical meanings. Every 20 steps of dynamics atoms that have ambiguous stereochemistry (like HB2/3) are swapped and the configuration with lowest energy is retained.
molecule :
file : input/1d3z.seq
type : nv
chain : 'A'
angles :
- file : input/dih.tbl
type : xplor
distances :
-
file : input/dis.tbl
type : xplor
-
file : input/hydcon.tbl
type : xplor
anneal:
dynOptions :
steps : 15000
highTemp : 5000.0
param :
swap : 20Here's an example of where the molecular structure and restraints are stored in a NEF file.
nef : input/1gb1.nef
anneal :
dynOptions :
steps : 15000
highTemp : 5000.0
param :
swap : 20Here's an example of .yaml file for an RNA where the sequence is explicitly listed in the .yaml file.
The indexing renumbers the sequence, in this case so it matches the numbering in the full virus sequence that this sequence is from.
The ptype parameter indicates that the sequence is RNA. In this example there are no explicit distance or angle restraints give in separate files.
Instead the vienna (dot-bracket) sequence is used to specify the secondary structure. Distance and angle restraints consistent with the helical (base-paired) region
are automaticallly generated. The ribose : Constrain value generate distance restraints to close up the ribose rings.
molecule :
entities :
- sequence : GGCUCUGGUGAGAGCCAGAGCC
indexing : '1:2 125:135 217:225'
ptype : RNA
rna:
ribose : Constrain
vienna : (((((((((....)))))))))
anneal:
dynOptions :
steps : 15000
highTemp : 5000.0
dfreeSteps : 0
force :
tors : 0.1
irp : -0.2Data files can be read in several formats.
NEF
First, molecular structure and constraints can be read together from an NMR Exchange Format (NEF) file. NEF files are a relatively recent format for storing NMR restraints. They are based on the STAR format that is used by both NMR-STAR (from the BMRB) and MMcif files. NMRFx can read NEF files to import the molecular sequence and distance, angle and RDC constraints. Indeed, one can generate a structure using default parameters by simply typing nmrfxs gen file.nef where file.nef is the name of the NEF file. If a nef file is used then it can be referenced in the .yaml file like:
nef : file.nefMolecular Topology
If you are not using NEF files, then you can read the molecular sequence and restraints from text files in a variety of formats. The molecular sequence can be read from a file in NMRViewJ (similar to CYANA) format. This is a file consisting of the three-letter amino acid code (one letter for RNA, and two letter for DNA). Each residue is on a single line of the file and can optionally be followed by an integer number representing the residue number. For example, a sequence starting with Methionine with sequential number 1 would look like:
MET 1
GLN
ILE
PHE4
VAL5
...If the sequence starts at 1 the integer residue position is optional. Any (or all) residue can have a sequential number next to it and gaps can be present in the numbering.
This file would be reference in the molecule section of the .yaml file:
molecule :
file : input/1d3z.seq
chain : A
type : nv
Alternatively the sequence can be explicitly listed as a single letter code in the .yaml file with a section like:
molecule :
entities :
- sequence : GGCUCUGGUGAGAGCCAGAGCC
indexing : '1:2 125:135 217:225'
ptype : RNAIn the above example the indexing field is used to specify the residue numbering. In this example the first residue would be 1, the second 2, the third 125. Further residues would be 126, 127 etc. up to position 135. Then the numbering continues with 217, 218 etc. through the last residue, 225.
Restraints
Restraints, as mentioned above, can be included in a NEF file along with the molecular topology. Alternatively they can be specified in several other formats. Currently supported are CYANA, XPLOR and an NMRFx format. Note: no attempt has been made to ensure comprehensive and fully accurate support of CYANA and XPLOR formats. They are provided as a convenience to the user and for our own testing. Our suite of test structures does, however, have a variety of examples of CYANA and XPLOR (and NEF) formats and all the examples run properly.
Here is the restraint section of a .yaml file (from our test suite) with CYANA restraints for ubiquitin. Note that CYANA distance restraints are normally in pairs with one for upper bounds (.upl) and one for lower bounds (.lol). The .yaml file need only mention the root name. So this example will load both distances.upl and distance.lol.
distances :
- file : input/distances
type : cyana
angles :
- file : input/dihedrals.aco
type : cyana
And the corresponding example with XPLOR restraints.
angles :
- file : input/dih.tbl
type : xplor
distances :
-
file : input/dis.tbl
type : xplor
-
file : input/hydcon.tbl
type : xplor
The NMRFx format has tab-separated lines with fields:
id, group, atom1Specifier, atom2Specifier, lower, upper
0 0 1:64.H 1:64.H 1.8 2.8
1 0 1:7.H 1:64.H 1.8 2.8
2 2 1:63.H 1:64.H 1.8 5.2
3 3 1:58.HA 1:64.H 1.8 4.3
4 3 1:64.HA 1:64.H 1.8 4.3There are a variety of parameters that effect the protocol used for simulated annealing in structure generation. The most likely ones that a user might want to change from the default value are steps, highTemp and cffSteps. Most other parameters can be left at their default values.
These parameters apply to the overall protocol and are set in the dynOptions section of the anneal section of the .yaml file.
The force terms apply during the calculation of energy values and gradients of the energy during commands such as gen. Generally, if a force term has a value < 0.0, then that component will not be included.
Structure calculation and refinement in NMRFX Structure proceeds through a series of stages that represent a process of simulated annealing. Each stage has preset values for Parameters and Forces. The values specify a static or falling temperature to be used during that stage as well as parameters that effect what atoms are included in the non-bond contacts and the forces used in the energy calculation. These preset values can be overwritten by entering values for any stage in the .yaml file. A normal structure calculation using the gen command proceeds through the following stages. The full set of parameters and force values for each stage can be seen below.
A final of gradient minimization is also included. The number of steps of this polishing phase is set by the polishSteps parameter. Otherwise this is not (yet) a full stage that can be modified like the other stages.
If the cffSteps parameter is set to a value greater than 0 then two additional stages will be inserted after the stage_anneal_low stage. These two stages, stage_cff_reduced and stage_cff_full are done with the non-bond contact forces calculated using a more realistic, but slower, calculation. These forces are based on parameters from the AMBER force field.
All the built-in stages can be displayed using the nmrfxs gen -y mode command which will print out all the stages in yaml format.
The mode argument specifies what stages are reported and can be gen, refine, cff or all. The output below was generated with:
nmrfxs gen -y allYou can also specify the name of a NEF file and it will be inserted into the output to give a complete project file that can be executed. For example to generate a full yaml file that will generate a structure based on data in test.nef you would use:
nmrfxs gen -y gen -n test.nef > project_test.yamland then generate a structure with
nmrfxs gen project_test.yamlNote: you don't need to generate a full yaml file for structure calculation as the normal gen command will use the default stages. But generating one as described here allows you to inspect and modify the values used in each stage.
Here's the output listing all available stages:
anneal:
dynOptions:
steps : 15000
highTemp : 5000.0
stage_prep:
param :
dislim : 4.6
swap : 20
updateAt : 5
hardSphere : 0.15
useh : False
shrinkValue : 0.2
ends : [3, 10, 20, 1000]
force :
dih : 5
irp : 0.05
dis : 1.0
repel : 0.5
stage_hi:
timestep : 4.0
dfreeSteps : None
switchFracVal : None
tempVal : 5000.0
econVal : 0.005
gMinSteps : None
param :
updateAt : 20
hardSphere : 0.15
useh : False
end : 1000
shrinkValue : 0.2
force :
dih : 5
dis : 1.0
repel : 0.5
nStepVal : 4500
stage_anneal_hi:
switchFracVal : None
tempVal : [5000.0, 250.0, 4.0]
econVal : 0.005
gMinSteps : None
param :
force :
nStepVal : 5200
stage_anneal_med:
switchFracVal : 0.65
tempVal : [250.0, 1.0, 4.0]
econVal : [0.005, 0.5]
gMinSteps : 100
param :
hardSphere : 0.0
useh : False
shrinkValue : 0.0
force :
nStepVal : 5200
stage_anneal_low:
switchFracVal : None
tempVal : None
econVal : None
gMinSteps : 100
param :
shrinkHValue : 0.0
hardSphere : 0.0
useh : True
shrinkValue : 0.0
force :
bondWt : 25.0
repel : 1.0
nStepVal : None
stage_cff_reduced:
switchFracVal : 0.2
tempVal : [100.0]
param :
dislim : 6.0
force :
nbmin : 1.0
dih : 5.0
stack : 0.1
irp : 0.5
tors : -0.1
dis : 40.0
repel : -1.0
cffnb : 1.0
nStepVal : 0
stage_cff_full:
switchFracVal : None
param :
force :
nbmin : 0.5
repel : -1.0
cffnb : 1.0
nStepVal : None
stage_low:
switchFracVal : None
tempVal : 0.0
econVal : 0.001
gMinSteps : 100
param :
force :
repel : 2.0
nStepVal : 100
A structure calculation with the gen subcommand generate several forms of output. During structure calculation information is written to standard output and is described below. At the completion of structure calculation three files are written into a folder named output (which is created if it doesn't already exist). These three files are tempN.pdb (PDB file of the generated structure), tempN.ang (dihedral angles of rotatable bonds in generated structure), and tempN.txt (angle and distance violations and close non-bond contacts). The N in these file names is the seed number used (specified with -s flag to the gen subcommand).
As described above the structure calculation proceeds through a series of stages. Each stage is characterized by a different set of Forces and Parameters and at the start of each stage the Forces and Parameters are reported. Each stage can involve both gradient minimization (typically at start of the stage) and rotational dynamics. The status of each of these protocols is periodically reported. Lines starting with GMIN are reported during gradient minimization and the output includes the current number of iterations, the number of non-bond contacts and the energy.
Note: at present, numbers for physics based parameters (time, kinetic energy, potential energy etc.) are not calibrated in a way that they are physically meaningful.
Lines starting with RDYN are reported during rotational dynamics.
Here is the output of the calculation of the structure of ubiquitin using data in the Example files.
nmrfxs gen project.yamlmass 8564.8
FORCES cffnb -1.00 nbmin 0.50 repel 0.50 elec -1.00 dis 1.00 dprob -1.00 dih 5.00 irp 0.05 shift -1.00 bondWt 1.00 stack -1.00
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 5 deltaStart 0 deltaEnd 2 disLim 4.60
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 5 deltaStart 0 deltaEnd 3 disLim 4.60
GMIN iter contacts energy
GMIN 0 2969 1251.01
GMIN 20 2389 2.77
GMIN 40 2421 -1.38
GMIN 60 2427 -1.73
GMIN 80 2431 -1.89
GMIN 84 2431 -1.90
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 5 deltaStart 0 deltaEnd 10 disLim 4.60
GMIN iter contacts energy
GMIN 0 3016 8600.56
GMIN 20 4307 207.02
GMIN 40 4049 123.20
GMIN 60 4019 99.19
GMIN 80 4004 91.30
GMIN 100 3971 88.14
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 5 deltaStart 0 deltaEnd 20 disLim 4.60
GMIN iter contacts energy
GMIN 0 6137 1231.51
GMIN 20 5264 169.32
GMIN 40 4870 135.81
GMIN 60 4760 124.97
GMIN 61 4760 124.97
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 5 deltaStart 0 deltaEnd 1000 disLim 4.60
GMIN iter contacts energy
GMIN 0 10386 10989.17
GMIN 20 7221 1160.59
GMIN 40 7026 749.82
GMIN 60 6895 628.65
GMIN 80 6795 600.66
GMIN 100 6279 439.39
FORCES cffnb -1.00 nbmin 0.50 repel 0.50 elec -1.00 dis 1.00 dprob -1.00 dih 5.00 irp 0.05 shift -1.00 bondWt 1.00 stack -1.00
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 20 deltaStart 0 deltaEnd 1000 disLim 4.60
init vel 5000.0 4999.999999999999
RDYN step time temp kinE potE totE deltaE timeStep rmsAngle maxAngle contacts
RDYN 0 0.000 4986.2 252.5 439.5 692.0 0.000000 0.000000 0.000 0.000 6290
RDYN 224 2908.659 4617.3 234.9 769.4 1004.3 0.007462 12.927374 1.703 6.469 5166
RDYN 449 4631.695 7885.3 387.3 433.9 821.2 0.004899 7.657937 1.221 4.583 4997
RDYN 674 7584.691 5314.4 263.9 610.0 873.9 0.006667 13.124428 2.034 6.996 6250
RDYN 899 9689.840 5171.8 278.5 503.7 782.1 0.005389 9.356217 1.544 5.482 5228
RDYN 1124 12177.481 5901.5 302.4 346.8 649.2 0.006183 11.056183 1.566 6.078 5229
RDYN 1349 15027.873 5291.9 269.0 673.5 942.5 0.007878 12.668408 2.154 7.802 6700
RDYN 1574 17785.140 5631.4 286.4 607.1 893.5 0.006683 12.254521 2.168 7.857 6442
RDYN 1799 20323.263 4740.6 242.2 503.8 746.0 0.007725 11.280546 2.020 7.575 5136
RDYN 2024 23152.798 5097.2 253.1 557.1 810.2 0.007492 12.575710 2.095 7.073 6087
RDYN 2249 25853.093 5964.4 291.0 524.9 816.0 0.006162 12.001310 1.808 6.832 5592
RDYN 2474 28249.377 5276.5 265.6 561.9 827.6 0.007235 10.650153 1.589 5.855 5132
RDYN 2699 30337.993 5189.4 266.9 516.1 783.0 0.005513 9.282740 1.613 6.068 5863
RDYN 2924 32649.062 5374.3 270.6 494.4 765.0 0.006198 10.271414 1.649 6.032 5211
RDYN 3149 35189.062 5252.3 264.4 715.4 979.7 0.006647 11.288890 1.923 6.707 7363
RDYN 3374 37311.469 5133.4 259.2 461.2 720.4 0.005736 9.432919 1.555 5.782 5604
RDYN 3599 40526.564 4426.2 229.6 614.7 844.3 0.006740 14.289312 2.052 7.699 6228
RDYN 3824 43259.732 5093.1 258.6 281.9 540.5 0.007406 12.147413 1.937 6.500 4765
RDYN 4049 45250.724 4978.3 254.1 462.0 716.1 0.005086 8.848856 1.341 5.192 5471
RDYN 4274 47776.677 5118.0 258.5 343.0 601.5 0.007157 11.226456 1.829 7.234 5278
RDYN 4499 50491.574 5133.2 259.5 592.3 851.8 0.005579 12.066210 1.942 6.799 4907
FORCES cffnb -1.00 nbmin 0.50 repel 0.50 elec -1.00 dis 1.00 dprob -1.00 dih 5.00 irp 0.05 shift -1.00 bondWt 1.00 stack -1.00
PARS coarse False includeH False hard 0.15 shrinkValue 0.20 shrinkHValue 0.00 updateAt 20 deltaStart 0 deltaEnd 1000 disLim 4.60
RDYN step time temp kinE potE totE deltaE timeStep rmsAngle maxAngle contacts
RDYN 0 50491.574 5124.5 259.5 616.5 876.0 0.000000 0.000000 0.000 0.000 4912
RDYN 259 53241.191 4334.8 246.0 324.2 570.3 0.004950 10.575451 1.603 6.059 4592
RDYN 519 56691.129 3451.4 171.6 290.5 462.1 0.006564 13.268990 1.533 5.702 4853
RDYN 779 60552.463 2801.3 146.2 227.0 373.2 0.005992 14.851288 1.645 6.401 4861
RDYN 1039 63514.610 2247.4 110.6 250.1 360.7 0.006071 11.392872 1.294 4.683 4282
RDYN 1299 66603.363 1695.2 86.6 142.7 229.4 0.005608 11.879820 1.309 4.413 4747
RDYN 1559 69369.128 1385.9 69.5 178.7 248.3 0.005656 10.637557 1.121 3.736 4580
RDYN 1819 72710.496 1051.3 53.5 139.0 192.5 0.005747 12.851416 1.218 4.157 4561
RDYN 2079 75827.206 863.8 43.4 102.4 145.8 0.005732 11.987345 0.974 3.361 4599
RDYN 2339 79647.275 708.6 35.5 79.4 115.0 0.005905 14.692575 1.142 4.083 4776
RDYN 2599 83460.276 560.5 28.2 64.0 92.3 0.006189 14.665387 0.997 3.604 4399
RDYN 2859 86517.756 446.1 22.6 47.8 70.4 0.005459 11.759540 0.783 3.043 4334
RDYN 3119 89579.102 385.3 19.6 38.4 58.0 0.006560 11.774405 0.723 2.816 4135
RDYN 3379 92890.222 325.5 16.4 29.2 45.6 0.005777 12.735077 0.703 2.919 4004
RDYN 3639 96200.934 295.9 15.0 39.7 54.6 0.005741 12.733508 0.682 2.548 4346
RDYN 3899 99842.233 262.4 13.3 30.8 44.1 0.005823 14.004999 0.688 2.460 4166
RDYN 4159 103709.173 255.5 12.9 29.8 42.7 0.005980 14.872846 0.727 2.701 4103
RDYN 4419 107622.884 239.1 12.0 35.4 47.4 0.006024 15.052734 0.722 3.119 4175
RDYN 4679 112043.490 242.1 12.2 28.2 40.4 0.006578 17.002332 0.798 3.921 4170
RDYN 4939 115432.651 256.8 13.1 20.1 33.2 0.005417 13.035234 0.592 2.137 4075
RDYN 5199 119050.572 255.9 13.0 25.0 38.0 0.005758 13.915081 0.631 2.397 4167
FORCES cffnb -1.00 nbmin 0.50 repel 0.50 elec -1.00 dis 1.00 dprob -1.00 dih 5.00 irp 0.05 shift -1.00 bondWt 1.00 stack -1.00
PARS coarse False includeH False hard 0.00 shrinkValue 0.00 shrinkHValue 0.00 updateAt 20 deltaStart 0 deltaEnd 1000 disLim 4.60
GMIN iter contacts energy
GMIN 0 4162 33.91
GMIN 20 3980 24.42
GMIN 33 3990 23.16
RDYN step time temp kinE potE totE deltaE timeStep rmsAngle maxAngle contacts
RDYN 0 119050.572 279.5 14.2 23.2 37.3 0.000000 0.000000 0.000 0.000 3991
RDYN 259 122208.906 200.1 10.2 31.8 42.0 0.005305 12.147438 0.524 1.911 3791
RDYN 519 125184.520 169.2 8.5 21.8 30.4 0.005121 11.444668 0.447 1.725 3820
RDYN 779 129409.778 132.9 6.7 15.3 22.1 0.005447 16.250994 0.593 2.078 3813
RDYN 1039 134108.677 107.6 5.4 15.1 20.6 0.005017 18.072686 0.577 2.035 3698
RDYN 1299 137968.677 80.7 4.1 12.9 17.0 0.005156 14.846157 0.419 1.588 3953
RDYN 1559 141240.646 62.8 3.2 8.9 12.1 0.004901 12.584495 0.318 1.343 3904
RDYN 1819 144177.171 46.5 2.4 3.1 5.4 0.005045 11.294328 0.250 0.842 3658
RDYN 2079 146403.431 34.5 1.7 1.6 3.3 0.004372 8.562537 0.172 0.642 3598
RDYN 2339 147651.564 24.1 1.2 1.0 2.2 0.004752 4.800511 0.084 0.314 3694
RDYN 2599 148096.764 16.8 0.8 -0.3 0.6 0.003911 1.712310 0.025 0.089 3601
RDYN 2859 148274.094 11.3 0.6 -0.5 0.0 0.004147 0.682038 0.008 0.032 3574
RDYN 3119 148312.818 7.5 0.4 -0.6 -0.2 0.004897 0.148939 0.001 0.004 3572
RDYN 3379 148424.632 4.8 0.2 -1.0 -0.7 0.003090 0.430052 0.003 0.009 3564
FORCES cffnb -1.00 nbmin 0.50 repel 1.00 elec -1.00 dis 1.00 dprob -1.00 dih 5.00 irp 0.05 shift -1.00 bondWt 25.00 stack -1.00
PARS coarse False includeH True hard 0.00 shrinkValue 0.00 shrinkHValue 0.00 updateAt 20 deltaStart 0 deltaEnd 1000 disLim 4.60
GMIN iter contacts energy
GMIN 0 17782 26.39
GMIN 20 16816 12.42
GMIN 40 16783 10.20
GMIN 60 16801 9.34
RDYN step time temp kinE potE totE deltaE timeStep rmsAngle maxAngle contacts
RDYN 3380 148424.632 7.8 0.4 9.3 9.7 0.000000 0.000000 0.000 0.000 16801
RDYN 3639 148734.054 3.3 0.2 4.3 4.4 0.002944 1.190086 0.008 0.046 16787
RDYN 3899 149036.606 2.1 0.1 1.9 2.0 0.003025 1.163663 0.007 0.051 16804
RDYN 4159 149336.382 1.5 0.1 0.8 0.9 0.002954 1.152981 0.005 0.032 16782
RDYN 4419 149547.240 1.2 0.1 0.3 0.4 0.002939 0.810992 0.003 0.015 16782
RDYN 4679 149664.833 1.0 0.1 0.1 0.2 0.002869 0.452281 0.002 0.008 16787
RDYN 4939 149724.268 1.0 0.1 0.0 0.1 0.002786 0.228596 0.001 0.004 16786
RDYN 5199 149753.403 1.0 0.1 -0.0 0.0 0.002690 0.112061 0.000 0.002 16794
FORCES cffnb -1.00 nbmin 0.50 repel 2.00 elec -1.00 dis 1.00 dprob -1.00 dih 5.00 irp 0.05 shift -1.00 bondWt 25.00 stack -1.00
PARS coarse False includeH True hard 0.00 shrinkValue 0.00 shrinkHValue 0.00 updateAt 20 deltaStart 0 deltaEnd 1000 disLim 4.60
GMIN iter contacts energy
GMIN 0 16794 4.18
GMIN 20 16683 3.71
GMIN 40 16698 3.47
GMIN 50 16691 3.46
RDYN step time temp kinE potE totE deltaE timeStep rmsAngle maxAngle contacts
RDYN 0 149753.403 1.0 0.1 3.5 3.5 0.000000 0.000000 0.000 0.000 16691
RDYN 4 149753.612 0.0 0.0 3.5 3.5 0.000001 0.041694 0.000 0.000 16691
RDYN 9 149753.848 0.0 0.0 3.5 3.5 0.000000 0.047173 0.000 0.000 16691
RDYN 14 149754.115 0.0 0.0 3.5 3.5 0.000000 0.053371 0.000 0.000 16691
RDYN 19 149754.358 0.0 0.0 2.8 2.8 0.050457 0.048579 0.000 0.000 16691
RDYN 24 149754.674 0.0 0.0 2.8 2.8 0.000000 0.063247 0.000 0.000 16691
RDYN 29 149755.032 0.0 0.0 2.8 2.8 0.000000 0.071558 0.000 0.000 16691
RDYN 34 149755.436 0.0 0.0 2.8 2.8 0.000000 0.080962 0.000 0.000 16691
RDYN 39 149755.894 0.0 0.0 2.8 2.8 0.000000 0.091601 0.000 0.000 16691
RDYN 44 149756.413 0.0 0.0 2.8 2.8 0.000000 0.103638 0.000 0.000 16691
RDYN 49 149756.999 0.0 0.0 2.8 2.8 0.000000 0.117257 0.000 0.000 16691
RDYN 54 149757.662 0.0 0.0 2.8 2.8 0.000000 0.132665 0.000 0.000 16691
RDYN 59 149758.413 0.0 0.0 2.8 2.8 0.000000 0.150098 0.000 0.000 16691
RDYN 64 149759.262 0.0 0.0 2.8 2.8 0.000000 0.169823 0.000 0.000 16691
RDYN 69 149760.223 0.0 0.0 2.8 2.8 0.000000 0.192139 0.000 0.000 16691
RDYN 74 149761.309 0.0 0.0 2.8 2.8 0.000001 0.217387 0.000 0.000 16691
RDYN 79 149762.539 0.0 0.0 2.8 2.8 0.000000 0.245954 0.000 0.000 16691
RDYN 84 149763.931 0.0 0.0 2.8 2.8 0.000001 0.278274 0.000 0.000 16691
RDYN 89 149765.505 0.0 0.0 2.8 2.8 0.000001 0.314841 0.000 0.000 16691
RDYN 94 149767.286 0.0 0.0 2.8 2.8 0.000000 0.356214 0.000 0.000 16691
RDYN 99 149769.301 0.0 0.0 2.8 2.8 0.000000 0.403024 0.000 0.000 16691
GMIN iter contacts energy
GMIN 0 16691 2.76
GMIN 6 16675 2.70
NNOE 7064 nRepel 16698
energy is 2.69516703654
etime 37.4120001793
Generation of multiple structures with the batch subcommand results in muliple simultaneous instances of the gen subcommand running. The output of each of these is redirected into files named cmdout_N.txt (where N is the seed number) in the output subdirectory.
Standard output of the batch command itself will be information about the progress of each calculation.
Lines like this indicate that a new NMRFx Structure process with the gen subcommand has been started and indicates the seed used, which of the requested number of structures this job represents, and the process identifier (pid) for the job.
submit 0 seed: 0 Structure # 1 of 50 pid 1509The number after submit indicates which of the reserved number of processes is being used for that job.
The batch process monitors spawned processes and when each one is completed it will print a Finish line:
Finish 1 seed: 1 Finished 1 of 50 pid 1511 eTime 56.2The batch process monitors each spawned processes elapsed time. If a process takes more than three times the average time of the jobs that have already finished then that job is killed. A new process (with new seed number) will be submitted.
Here's part of the output for generation of 50 structures:
nmrfxs batch -n 50 -k 10 -p 5 -a -c project.yaml
submit 0 seed: 0 Structure # 1 of 50 pid 1509
submit 1 seed: 1 Structure # 2 of 50 pid 1511
submit 2 seed: 2 Structure # 3 of 50 pid 1515
submit 3 seed: 3 Structure # 4 of 50 pid 1518
submit 4 seed: 4 Structure # 5 of 50 pid 1523
Finish 1 seed: 1 Finished 1 of 50 pid 1511 eTime 56.2
Finish 2 seed: 2 Finished 2 of 50 pid 1515 eTime 56.2
submit 1 seed: 5 Structure # 6 of 50 pid 1568
submit 2 seed: 6 Structure # 7 of 50 pid 1569
Finish 0 seed: 0 Finished 3 of 50 pid 1509 eTime 57.2
Finish 3 seed: 3 Finished 4 of 50 pid 1518 eTime 57.2
Finish 4 seed: 4 Finished 5 of 50 pid 1523 eTime 57.2
submit 0 seed: 7 Structure # 8 of 50 pid 1580
submit 3 seed: 8 Structure # 9 of 50 pid 1581
submit 4 seed: 9 Structure # 10 of 50 pid 1584
...
submit 0 seed: 47 Structure # 48 of 50 pid 2005
submit 4 seed: 48 Structure # 49 of 50 pid 2006
Finish 2 seed: 44 Finished 45 of 50 pid 1965 eTime 57.1
submit 2 seed: 49 Structure # 50 of 50 pid 2020
Finish 3 seed: 45 Finished 46 of 50 pid 1991 eTime 57.1
Finish 1 seed: 46 Finished 47 of 50 pid 1999 eTime 55.1
Finish 0 seed: 47 Finished 48 of 50 pid 2005 eTime 55.1
Finish 2 seed: 49 Finished 49 of 50 pid 2020 eTime 55.1
Finish 4 seed: 48 Finished 50 of 50 pid 2006 eTime 56.1
Done
* ca,c,n,o,p,o5',c5',c4',c3',o3'
repModel 6 rms 0.971581149907 avgrms 1.13736317587
coreResidues [u'A:2-7', u'A:9-72']
coreResidues [u'A:2-7', u'A:11-71']
coreResidues [u'A:2-6', u'A:11-71']
coreResidues [u'A:2-6', u'A:11-71']
coreResidues [u'A:2-6', u'A:11-71']
repModel 3 rms 0.464075104455 avgrms 0.509412597497When all requested structures are finished the output for each structure calculation will be analyzed and the N (where N is set with the -k flag) best structures will be kept. The best structures (and their corresponding angle and violation files) will be copied from output to the final directory. They will be given extensions starting with 1, for the best structure (final1.pdb, final1.ang, final1.txt).
The super subcommand will be used to generate an MMCif file with all the superimposed structures or a set of pdb files (sup_final1.pdb etc). The choice of MMcif (default) or PDB file is set by the -t flag to the batch command.
The summary subcommand will be used to generate an analysis.txt file in the final directory. It will contain distance violations and close contacts that have distance violations greater than a specified amount (default 0.2) and are present in more than a specified (default 2) structures:
Type Atom1 Atom2 nViol Bound Mean Max Structures
Dis: 1:50.HA - 1:50.HD21 3 2.55 0.12 0.45 |..++..+...|
Dis: 1:50.HA - 1:50.HD22 3 2.55 0.12 0.45 |..++..+...|
Dis: 1:50.HA - 1:50.HD23 3 2.55 0.12 0.45 |..++..+...|A summary of energy terms from the last line of the violation files (final1.txt, final2.txt etc.) will be written to final/summary.txt. Each line of the summary corresponds to one final structure. The value in the first column is the index of the file (1 for final1.pdb etc.). The second column (34, 36, 15 etc. in example below) is the seed used to generate that structure.
1 34 Irp 307 -4.380 Dih 97 0.208 CFF 0 0.000 Repel 16565 4.692 Distance 7064 1.356 0.450 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 1.875
2 36 Irp 307 -4.122 Dih 97 0.343 CFF 0 0.000 Repel 17172 4.596 Distance 7064 1.323 -0.328 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.139
3 15 Irp 307 -4.502 Dih 97 0.246 CFF 0 0.000 Repel 17070 4.668 Distance 7064 1.799 -0.470 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.210
4 37 Irp 307 -4.349 Dih 97 0.323 CFF 0 0.000 Repel 17097 4.387 Distance 7064 1.860 0.455 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.221
5 10 Irp 307 -4.349 Dih 97 0.115 CFF 0 0.000 Repel 17043 4.500 Distance 7064 1.979 0.455 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.244
6 12 Irp 307 -4.430 Dih 97 0.250 CFF 0 0.000 Repel 17171 4.495 Distance 7064 1.951 0.350 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.266
7 21 Irp 307 -4.331 Dih 97 0.113 CFF 0 0.000 Repel 16955 4.895 Distance 7064 1.597 0.452 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.274
8 38 Irp 307 -4.534 Dih 97 0.275 CFF 0 0.000 Repel 17102 4.727 Distance 7064 1.891 -0.441 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.359
9 41 Irp 307 -4.566 Dih 97 0.128 CFF 0 0.000 Repel 17386 4.865 Distance 7064 1.932 0.439 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.360
10 35 Irp 307 -4.363 Dih 97 0.138 CFF 0 0.000 Repel 17419 4.774 Distance 7064 1.872 0.515 Shift 0 0.000 ProbT 0 0.000 Stack 0 0.000 Total 2.421