IMODfit User Guide

Here we give a brief overview of the necessary steps to run the program but we strongly encourage to follow the tutorials. The user guide is partitioned in three levels of usage: 

Minimal options

 For flexible fitting a single atomic structure into a EM density map enter the following command at the prompt.

imodfit <pdb> <map> <resolution> <cutoff>


pdb PDB initial model (required).
map Target EM map file (required). Situs and CCP4 formats are accepted; but extensions ".sit" and ".cpp4" are mandatory.
resolution Resolution (required). Our resolution criterion follows EMAN package procedures (e.g. PDB2MRC tool)
Cutoff EM map threshold (required). All density levels below this value will not be considered.

iMODFIT outputs the following files: 

  • imodfit_fitted.pdb --> fitted  atomic structure
  • imodfit_score.txt --> score file to check for convergence
  • imodfit.log--> used command log

Basic Options

 The more common options to customize your fitting are:

−h Displays usage information and exits
−i <int> Maximum number of iterations (default=10000).
−m Coarse-Grained model: 0=CA, 1=3BB2R, 2=Full-Atom, 3=NCAC(experimental) (default=2).
−x Considers first CHI dihedral angle (default=disabled).
−t Outputs a Multi-PDB trajectory movie (<basename_movie>.pdb)
−n <int/float> Used modes range, either number [1,N] <integer>, or ratio [0,1) <float> (default=0.2).
−e <int/float> Excited modes range, either number [1,nevs] <integer>, or ratio [0,1) <float> (default=0.02)
−F Enables full-atom output models.
−o <string> Output files basename (default=imodfit).
−P <int> Pairwise interaction potential: (default=0)
    0=inverse exponential (= k/(1+(x/x0)^p), if x < c, else k=0)
    1=cutoff (= k, if x < c, else k=0)
    2=Topology & Secondary Structure (--func is mandatory)
By default an extra torsional potential will be added.
−f <string> ASCII file defining the ICs to be fixed with the format:
    Protein:       "index phi chi psi"
    NAcid:         "index alpha beta gamma chi epsilon zeta"
    Inter-chain:   "index IC"
Where IC-name= 0(fixed) or 1(mobile)
A demo file can be generated using iMODE with the --save_fixfile option.
−r <float> Randomly fixed ratio of Dihedral Coords. Example: 0.7 = 70% of dihedrals will be randomly fixed. (Rotational/Translational coords. always mobile)
−S All dihedral coordinates belonging to residues matching the indicated SS identifieres will be fixed. Ex: "HE" will fix the dihedrals corresponding to alpha-helices and beta-sheets.

Advanced options

Only for real expert users!

   --ss <string>
      Secondary Structure ASCII file with 2 cols.: <index> <char>
     Where <index> is the corresponding residue index (1,2,...), and <char>
     is the single character SS identifier. By default SS will be computed
     internally (H=helix, E=strand, C=coil). 
   --func <string>
      ASCII file defining the force constant functions to be applied
     according to Topology and/or Secondary Structure. The 5 cols. format
     is: <SS> <n> <k> <x0> <pow>
     Where <SS> is the two character pairwise interaction identifier, <n>
     is the topology, and <k>,<x0>,<pow> are the corresponding inverse
     exponential function parameters. If --ss is not specified, only the XX
     pairwise interaction identifier will be considered. If <n> is 0, any
     previously not-matched topology will be matched. 
   --k0_c <float>
      Inverse Exponential's distance cutoff (default=10A) 
   --k0_k <float>
      Inverse Exponential's stiffness constant (default=1.0) 
   --k0_x0 <float>
      Inverse Exponential's inflexion point (default=3.8A) 
   --k0_p <float>
      Inverse Exponential's power term (default=6) 
   --k1_c <float>
      Distance cutoff method distance cutoff (default=10A) 
   --k1_k <float>
      Distance cutoff method stiffness constant (default=1.0) 
   --k2_c <float>
      Non-bonding distance cutoff applied to --ss_func (default=10A) 
   --pdb_ref <string>
      Reference PDB file (test only) 
      Enable initial (basename_model.pdb) pdb model output
      Disables mass weighting (default=disabled) 
      Disables extra torsional potential (default=disabled) 
      Disables Gaussian Weighted RMSD (default=disabled) 
   --cutoff2 <float>
      Density cutoff of simulated map (default = 0.001) 
   --filter <int>
      Select filtration method: 1-Fourier, 2-Kernel (By default the fastest
     will be selected) 
   --6Dref <int>
      Number of iterations between local 6D pose refinement (default=200) 
   --6Dref_delay <int>
      Number of iterations before first local 6D pose refinement
      Independent local 6D pose refinement for each single chain. 
   --6Dref_trans <float>
      Maximum translational increment (Angstroms) for pose refinement.
   --6Dref_rot <float>
      Maximum rotational increment (degrees) for pose refinement
   --prob <string>
      Normal Mode Selection Probabitity. --nex modes will be selected and
     merged from --nevs subset according to the following probabilities
     (p): (default=var)
     	plain: equiprobability (--min_prob will be ignored)
     	var: proportional to the mode variance, p(i)= 1/eigenvalue(i)
     	line: lineally decreasing probability, p(i)= 1-i/nevs 
   --rediag <double>
      RMSD ratio to trigger diagonalization.
     (last_diag_RMSD-current_RMSD)/first_RMSD > rediag (default=0.05) 
   --delta_save <float>
      RMSD increment from previous saved frame to trigger frame saving.
   --wrmsd <float>
      Sets the RMSD weighting [0:1] during model alingment (wrmsd=0, means
     maximum weighting) (default=0) 
   --conv <double>
      Convergence threshold (relative to first score) (default=1e-8) 
   --conv_win <int>
      Window length (number of iterations) to estimate convergence.
   --seed <unsigned int>
      Pre-define the random number generator SEED (Mersenne Twister)
     (default=random-seed from /dev/urandom) 
   --verb <int>
      Verbosity display level (0=low, 1=medium, 2=high) (default=0) 
   --step <double>
      Initial amplitude applied to excited modes (default=10) 
   --step2 <double>
      Final amplitude applied to excited modes (amplitude will decrease
     lineally from "step" to "step2") (default=step/10) 
      Random amplitude selection between step and step2 (default=disabled) 
   --nex2 <int>
      Final number of excited eigenvectors, either number [1,nevs]
     <integer>, or ratio [0,1) <float>. Number of excited modes will change
     lineally from "nevs" to "nevs2" (default=disabled) 
      Random number of excited modes between nex and nex2
      Disables random excited modes weighting (default=disabled) 
   --,  --ignore_rest
      Ignores the rest of the labeled arguments following this flag. 
   -v,  --version
      Displays version information and exits.