Difference between revisions of "Modes"

From Phaserwiki
m (moved Modes to Keyword Modes: consistency with python)
(Normal Mode Analysis/SCEDS)
 
(63 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
<div style="margin-left: 25px; float: right;">__TOC__</div>
 
<div style="margin-left: 25px; float: right;">__TOC__</div>
[[Category: Phaser-2.2]]
 
==Modes==
 
  
Phaser runs in different modes, which perform Phaser's different functionalities.  Modes can either be basic modes or modes that combine the functionality of basic modes.  The mode of operation is controlled with the MODE keyword.
+
Phaser runs in different modes, which perform Phaser's different functionalities.  Modes can either be basic modes or modes that combine the functionality of basic modes.   
  
*'''Automated Molecular Replacement'''
+
===Phaser Executable===
*: Mode MR_AUTO combines anisotropy correction, cell content analysis, fast rotation and translation functions, and refinement and phasing to automatically solve a structure by molecular replacement
 
*'''Fast Rotation Function'''
 
*: Mode MR_FRF combines the anisotropy correction and likelihood-enhanced fast rotation function, by default rescored with the full rotation likelihood function,.
 
*'''Fast Translation Function'''
 
*: Mode MR_FTF combines the anisotropy correction and likelihood-enhanced fast translation function, by default rescored by the full likelihood translation function.
 
*'''Brute Rotation Function'''
 
*: Mode MR_BRF combines the anisotropy correction and a brute-force search of angles using the full likelihood rotation function.
 
*'''Brute Translation Function'''
 
*: Mode MR_BTF combines the anisotropy correction and a brute force search of positions using the full likelihood translation function to find the position of a previously oriented model.
 
*'''Packing'''
 
*: Mode MR_PAK determines whether molecular replacement solutions pack in the unit cell using a C-alpha clash test.
 
*'''Log-Likelihood Gain'''
 
*: Mode MR_LLG combines the anisotropy correction and the likelihood function to calculate the log-likelihood gain for full or partial molecular replacement solutions.
 
*'''Refinement and Phasing'''
 
*: Mode MR_RNP combines the anisotropy correction and refinement against the likelihood function to optimize full or partial molecular replacement solutions and phase the data. At the end of refinement, the list of solutions is checked for duplicates, which are pruned.
 
*'''Anisotropy Correction'''
 
*: Mode ANO corrects the experimental data (amplitude and associated sigma) for anisotropy. First, the anisotropy is removed by scaling up data from weak directions and scaling down data from strong directions, preserving the overall equivalent isotropic B-factor.  Then (as suggested by Strong ''et al.'' (2006), PNAS 103:8060-8065), the data are resharpened to restore the original falloff in the strong direction.  The amount of resharpening can be controlled with the RESHARP keyword.
 
*'''Cell Content Analysis'''
 
*: Mode CCA determines the composition of the crystals using the "new" Matthews coefficients of Kantardjieff & Rupp (2003) ("Matthews coefficient probabilities: Improved estimates for unit cell contents of proteins, DNA and protein-nucleic acid complex crystals". Protein Science 12:1865-1871). The molecular weight of ONE complex or assembly to be packed into the asymmetric unit is given and the possible Z values (number of copies of the complex or assembly) that will fit in the asymmetric unit and the relative frequency of their corresponding VM values is reported.
 
*'''Normal Mode Analysis'''
 
*: Mode NMA writes out pdb files that have been perturbed along normal modes, in a procedure similar to that described by Suhre & Sanejouand (Acta Cryst. D60, 796-799, 2004). Each run of the program writes out a matrix FILEROOT.mat that contains the eigenvectors and eigenvalues of the atomic Hessian, and can be read into subsequent runs of the same job, to speed up the analysis.
 
*'''Automated Experimental Phasing'''
 
*: Mode EP_AUTO combines anisotropy correction, cell content analysis, and SAD phasing to automatically solve a structure by experimental phasing
 
*'''SAD Phasing'''
 
*: Mode EP_SAD refines atoms using the SAD likelihood function, and completes the structure with log-Likelihood gradient maps.
 
  
==Keywords for Modes==
+
The Phaser executable runs in different modes, which perform Phaser's different functionalities. The mode is selected with the [[Keywords#MODE | MODE]] keyword.
  
 +
===Python Interface===
 +
Phaser can be compiled as a python library. The mode is selected by calling the appropriate run-job. Input to the run-job is via input-objects, which are passed to the run-job. Setter function on the input objects are equivalent to the keywords for input to the phaser executable. See [[Python Interface]] for details.
  
{| class="wikitable"
+
===Modes===
|-
+
 
!Mode !! Compulsory  !!Optional !!Expert
+
====Automated Molecular Replacement====
|-
+
:; MODE MR_AUTO  
! MR_AUTO  
+
:Combines cell content analysis, anisotropy correction, translational NCS analysis, rotation function, [optional rotation refinement], translation function, packing, and refinement to automatically solve a structure by molecular replacement
| ENSEMBLE HKLIN LABIN MODE SEARCH
+
:Searches for multiple components by looping through the steps listed above
|| COMPOSITION HKLOUT FINAL HKLOUT PACK PERMUTATIONS PURGE RESCORE RESOLUTION ROOT  SCRIPT SGALTERNATIVE SOLUTION SPACEGROUP TITLE TOPFILES VERBOSE XYZOUT
+
:Two different search methods are available
|| BINS BOXS CELL CLMN MUTE OUTLIER SAMPLING SOLPARAMS
+
:: '''FAST''' ''Uses amalgamation to combine placements from a single translation function''
|-
+
:: '''FULL''' ''Heavily branched search without amalgamation'''
! MR_FRF
+
  ResultMR r = runMR_AUTO(i=InputMR_AUTO)
| ENSEMBLE HKLIN LABIN MODE SEARCH
+
 
|| COMPOSITION FINAL RESCORE RESOLUTION ROOT  SCRIPT  SOLUTION SPACEGROUP TITLE TOPFILES VERBOSE XYZOUT
+
====Rotation Function====
|| BINS BOXS CELL CLMN MUTE OUTLIER SAMPLING SOLPARAMS TARGET
+
:; MODE MR_ROT
|-
+
:Combines the anisotropy correction, translational NCS analysis, and EITHER likelihood-enhanced fast rotation function, by default rescored with the full rotation likelihood function OR a brute-force search of angles using the full likelihood rotation function
! MR_BRF
+
  ResultMR_RF r = runMR_FRF(i=InputMR_FRF)
| ENSEMBLE HKLIN LABIN MODE SEARCH
+
 
|| COMPOSITION FINAL RESOLUTION ROOT ROTATION SCRIPT  SOLUTION SPACEGROUP TITLE TOPFILES VERBOSE XYZOUT
+
====Translation Function====
|| BINS BOXS CELL MUTE OUTLIER SAMPLING SOLPARAMS
+
:; MODE MR_TRA
|-
+
:Combines the anisotropy correction, translational NCS analysis, and EITHER likelihood-enhanced fast translation function, by default rescored by the full likelihood translation function OR a brute force search of positions using the full likelihood translation function to find the position of a previously oriented model
! MR_FTF
+
  ResultMR_TF r = runMR_FTF(i=InputMR_FTF)
| ENSEMBLE HKLIN LABIN MODE SOLUTION
+
 
|| COMPOSITION FINAL RESCORE RESOLUTION ROOT SCRIPT SGALTERNATIVE SPACEGROUP TITLE TOPFILES VERBOSE XYZOUT
+
====Packing====
|| BINS BOXS CELL MUTE OUTLIER SAMPLING SOLPARAMS TARGET
+
:; MODE MR_PAK
|-
+
:Determines whether molecular replacement solutions pack in the unit cell using a C-alpha clash test
! MR_BTF
+
  ResultMR r = runMR_PAK(i=InputMR_PAK)
| ENSEMBLE HKLIN LABIN MODE SOLUTION
+
 
|| COMPOSITION FINAL RESOLUTION ROOT SCRIPT SGALTERNATIVE  SPACEGROUP TITLE TOPFILES TRANSLATE VERBOSE XYZOUT
+
====Refinement and Phasing====
|| BINS BOXS CELL MUTE OUTLIER SAMPLING SOLPARAMS
+
:; MODE MR_RNP
|-
+
:Combines the anisotropy correction, translational NCS analysis, and refinement against the likelihood function to optimize full or partial molecular replacement solutions and phase the data. At the end of refinement, the list of solutions is checked for duplicates, which are pruned
! MR_RNP
+
  ResultMR r = runMR_RNP(i=InputMR_RNP)
| ENSEMBLE HKLIN LABIN MODE SOLUTION
+
:; MODE [http://www.jabberwocky.com/carroll/jabber/jabberwocky.html GYRE]
|| COMPOSITION HKLOUT RESOLUTION ROOT SCRIPT SPACEGROUP TITLE TOPFILES VERBOSE XYZOUT
+
:Combines the anisotropy correction, translational NCS analysis, and refinement against the likelihood function to optimize the rotations and relative translations (gyrations) between chains in an ensemble
|| BINS BOXS CELL MACMR MUTE OUTLIER SOLPARAMS
+
  ResultGYRE r = runMR_GYRE(i=InputMR_RNP)
|-
+
:; MODE [http://www.jabberwocky.com/carroll/jabber/jabberwocky.html GIMBLE]
! MR_LLG
+
:Combines the anisotropy correction, translational NCS analysis, and refinement against the likelihood function to optimize the rotations and translations between chains in an ensemble
| ENSEMBLE HKLIN LABIN MODE SOLUTION
+
  ResultMR r = runMR_GMBL(i=InputMR_RNP)
|| COMPOSITION HKLOUT RESOLUTION ROOT SCRIPT SPACEGROUP TITLE TOPFILES VERBOSE XYZOUT
+
:; MODE PRUNE
|| BINS BOXS CELL MUTE OUTLIER SOLPARAMS
+
:Combines the anisotropy correction, translational NCS analysis, and occupancy refinement against the likelihood function to optimize the occupancies of residues. Occupancy refinement is performed for occupancies in sliding residue windows of a size determined by the expected LLG. The resides that decrease the LLG are pruned.
|-
+
  ResultMR r = runMR_OCC(i=InputMR_OCC)
! MR_PAK
+
 
| ENSEMBLE HKLIN LABIN MODE SOLUTION
+
====Anisotropy Correction====
|| PACK ROOT SCRIPT SPACEGROUP TITLE VERBOSE TOPFILES XYZOUT
+
:; MODE ANO
|| CELL MUTE
+
:Corrects the experimental data (amplitude and associated sigma) for anisotropy. First, the anisotropy is removed by scaling up data from weak directions and scaling down data from strong directions, preserving the overall equivalent isotropic B-factor.  Then (as suggested by Strong ''et al.'' (2006), PNAS 103:8060-8065), the data are resharpened to restore the original falloff in the strong direction.  The amount of resharpening can be controlled with the RESHARP keyword.
|-
+
  ResultANO r = runANO(i=InputANO)
! EP_AUTO
+
 
| ATOM CRYSTAL HKLIN LABIN MODE
+
====Translational NCS and Twin analysis====
|| COMPOSITION HAND HKLOUT LLGCOMPLETE ROOT SCRIPT SPACEGROUP TITLE VERBOSE XYZOUT
+
:; MODE NCS
|| BFACTOR CELL FFTS MACSAD MUTE OUTLIER VARSAD
+
:Finds pseudo-translational NCS vectors and corrects the data for intensity variations due to the ptNCS using likelihood methods. The data are corrected for anisotropy first and analysed for twinning before and after data correction.
|-
+
  ResultNCS r = runNCS(i=InputNCS)
! EP_SAD
+
 
| ATOM CRYSTAL HKLIN LABIN MODE
+
====Cell Content Analysis====
|| HAND HKLOUT LLGCOMPLETE PART ROOT SCRIPT SPACEGROUP TITLE VERBOSE XYZOUT
+
:; MODE CCA
|| BFACTOR CELL FFTS MACSAD MUTE OUTLIER VARSAD
+
:Determines the composition of the crystals using the "new" Matthews coefficients of Kantardjieff & Rupp (2003) ("Matthews coefficient probabilities: Improved estimates for unit cell contents of proteins, DNA and protein-nucleic acid complex crystals". Protein Science 12:1865-1871). The molecular weight of ONE complex or assembly to be packed into the asymmetric unit is given and the possible Z values (number of copies of the complex or assembly) that will fit in the asymmetric unit and the relative frequency of their corresponding VM values is reported.
|-
+
  ResultCCA r = runCCA(i=InputCCA)
! ANO
+
 
| HKLIN LABIN MODE
+
====Expected log-likelihood-gain (eLLG) Analysis====
|| ROOT SCRIPT SPACEGROUP TITLE VERBOSE
+
:; MODE MR_ELLG
|| CELL MACANO MUTE
+
:Determines the eLLG for any models defined as ensembles, reporting the eLLG at the full resolution of the data and the resolution predicted to reach the target LLG.
|-
+
  ResultELLG r = runMR_ELLG(i=InputMR_ELLG)
! CCA
+
 
| COMPOSITION HKLIN LABIN MODE
+
====Normal Mode Analysis/SCEDS====
|| RESOLUTION ROOT SCRIPT TITLE VERBOSE
+
:; MODE NMA
|| CELL MUTE
+
:Writes out coordinates of input structures that have been perturbed along normal modes, in a procedure similar to that described by Suhre & Sanejouand (Acta Cryst. D60, 796-799, 2004). Perturbed coordinates are output in RMS increments for use in MR trials. Each run of the program writes out a matrix FILEROOT.mat that contains the eigenvectors and eigenvalues of the atomic Hessian, and can be read into subsequent runs of the same job, to speed up the analysis.
|-
+
  ResultNMA r = runNMAXYZ(i=InputNMA)
! NMA
+
 
| ENSEMBLE MODE
+
:; MODE SCEDS
|| EIGEN NMAMETHOD NMAPDB ROOT SCRIPT TITLE VERBOSE
+
:Writes out domains generated by the SCEDS analysis of the coordinates perturbed along normal modes.
|| MUTE
+
  ResultNMA r = runSCEDS(i=InputNMA)
|}
+
 
 +
====Automated Experimental Phasing====
 +
:; MODE EP_AUTO
 +
:Combines anisotropy correction, translational NCS analysis, cell content analysis, and SAD phasing to automatically solve a structure by experimental phasing
 +
  ResultEP r = runEP_AUTO(i=InputEP_AUTO)
 +
====Single Atom Molecular Replacement====
 +
:; MODE MR_ATOM
 +
:Combines automated molecular replacement with log-likelihood gradient completion to solve high resolution structures by single atom MR. The data are prepared with runMR_DAT and then the phasing is done withing the EP_AUTO mode.
 +
  ResultEP r = runMR_ATOM(i=InputMR_ATOM)

Latest revision as of 12:23, 5 September 2016

Phaser runs in different modes, which perform Phaser's different functionalities. Modes can either be basic modes or modes that combine the functionality of basic modes.

Phaser Executable

The Phaser executable runs in different modes, which perform Phaser's different functionalities. The mode is selected with the MODE keyword.

Python Interface

Phaser can be compiled as a python library. The mode is selected by calling the appropriate run-job. Input to the run-job is via input-objects, which are passed to the run-job. Setter function on the input objects are equivalent to the keywords for input to the phaser executable. See Python Interface for details.

Modes

Automated Molecular Replacement

MODE MR_AUTO
Combines cell content analysis, anisotropy correction, translational NCS analysis, rotation function, [optional rotation refinement], translation function, packing, and refinement to automatically solve a structure by molecular replacement
Searches for multiple components by looping through the steps listed above
Two different search methods are available
FAST Uses amalgamation to combine placements from a single translation function
FULL' Heavily branched search without amalgamation
 ResultMR r = runMR_AUTO(i=InputMR_AUTO)

Rotation Function

MODE MR_ROT
Combines the anisotropy correction, translational NCS analysis, and EITHER likelihood-enhanced fast rotation function, by default rescored with the full rotation likelihood function OR a brute-force search of angles using the full likelihood rotation function
 ResultMR_RF r = runMR_FRF(i=InputMR_FRF)

Translation Function

MODE MR_TRA
Combines the anisotropy correction, translational NCS analysis, and EITHER likelihood-enhanced fast translation function, by default rescored by the full likelihood translation function OR a brute force search of positions using the full likelihood translation function to find the position of a previously oriented model
 ResultMR_TF r = runMR_FTF(i=InputMR_FTF)

Packing

MODE MR_PAK
Determines whether molecular replacement solutions pack in the unit cell using a C-alpha clash test
 ResultMR r = runMR_PAK(i=InputMR_PAK)

Refinement and Phasing

MODE MR_RNP
Combines the anisotropy correction, translational NCS analysis, and refinement against the likelihood function to optimize full or partial molecular replacement solutions and phase the data. At the end of refinement, the list of solutions is checked for duplicates, which are pruned
 ResultMR r = runMR_RNP(i=InputMR_RNP)
MODE GYRE
Combines the anisotropy correction, translational NCS analysis, and refinement against the likelihood function to optimize the rotations and relative translations (gyrations) between chains in an ensemble
 ResultGYRE r = runMR_GYRE(i=InputMR_RNP)
MODE GIMBLE
Combines the anisotropy correction, translational NCS analysis, and refinement against the likelihood function to optimize the rotations and translations between chains in an ensemble
 ResultMR r = runMR_GMBL(i=InputMR_RNP)
MODE PRUNE
Combines the anisotropy correction, translational NCS analysis, and occupancy refinement against the likelihood function to optimize the occupancies of residues. Occupancy refinement is performed for occupancies in sliding residue windows of a size determined by the expected LLG. The resides that decrease the LLG are pruned.
 ResultMR r = runMR_OCC(i=InputMR_OCC)

Anisotropy Correction

MODE ANO
Corrects the experimental data (amplitude and associated sigma) for anisotropy. First, the anisotropy is removed by scaling up data from weak directions and scaling down data from strong directions, preserving the overall equivalent isotropic B-factor. Then (as suggested by Strong et al. (2006), PNAS 103:8060-8065), the data are resharpened to restore the original falloff in the strong direction. The amount of resharpening can be controlled with the RESHARP keyword.
 ResultANO r = runANO(i=InputANO)

Translational NCS and Twin analysis

MODE NCS
Finds pseudo-translational NCS vectors and corrects the data for intensity variations due to the ptNCS using likelihood methods. The data are corrected for anisotropy first and analysed for twinning before and after data correction.
 ResultNCS r = runNCS(i=InputNCS)

Cell Content Analysis

MODE CCA
Determines the composition of the crystals using the "new" Matthews coefficients of Kantardjieff & Rupp (2003) ("Matthews coefficient probabilities: Improved estimates for unit cell contents of proteins, DNA and protein-nucleic acid complex crystals". Protein Science 12:1865-1871). The molecular weight of ONE complex or assembly to be packed into the asymmetric unit is given and the possible Z values (number of copies of the complex or assembly) that will fit in the asymmetric unit and the relative frequency of their corresponding VM values is reported.
 ResultCCA r = runCCA(i=InputCCA)

Expected log-likelihood-gain (eLLG) Analysis

MODE MR_ELLG
Determines the eLLG for any models defined as ensembles, reporting the eLLG at the full resolution of the data and the resolution predicted to reach the target LLG.
 ResultELLG r = runMR_ELLG(i=InputMR_ELLG)

Normal Mode Analysis/SCEDS

MODE NMA
Writes out coordinates of input structures that have been perturbed along normal modes, in a procedure similar to that described by Suhre & Sanejouand (Acta Cryst. D60, 796-799, 2004). Perturbed coordinates are output in RMS increments for use in MR trials. Each run of the program writes out a matrix FILEROOT.mat that contains the eigenvectors and eigenvalues of the atomic Hessian, and can be read into subsequent runs of the same job, to speed up the analysis.
 ResultNMA r = runNMAXYZ(i=InputNMA)
MODE SCEDS
Writes out domains generated by the SCEDS analysis of the coordinates perturbed along normal modes.
 ResultNMA r = runSCEDS(i=InputNMA)

Automated Experimental Phasing

MODE EP_AUTO
Combines anisotropy correction, translational NCS analysis, cell content analysis, and SAD phasing to automatically solve a structure by experimental phasing
 ResultEP r = runEP_AUTO(i=InputEP_AUTO)

Single Atom Molecular Replacement

MODE MR_ATOM
Combines automated molecular replacement with log-likelihood gradient completion to solve high resolution structures by single atom MR. The data are prepared with runMR_DAT and then the phasing is done withing the EP_AUTO mode.
 ResultEP r = runMR_ATOM(i=InputMR_ATOM)