Contents
Introduction
Release
History
Contact
Faq
Overview
Conventions
Abbreviations
Intro
Reference Guide
User's Guide
References
Glossary
Index

[ Faq ]

 
 https://www.molsoft.com/support 

 
 support@molsoft.com 

[ Faq kernel | Faq ellipsoid | Faq residue properties ]

Trouble-shooting. Setup on LINUX

ICM does not start

Q: ICM does not start on a new Linux installation. Is there a solution?

A:ICM is compiled on a computer with the the old Linux kernel for compatibility with the old kernel. As a result, you may get this message when you start ICM:

icm
 icm: relocation error: icm: symbol errno, version GLIBC_2.0 not defined 
 in file libc.so.6 with link time reference

To run this executable on a new kernel, set the LD_ASSUME_KERNEL shell variable to 2.4.21

If you use the same executable in a heterogenious kernel environment, you may create a shell script to start ICM in which the LD_ASSUME_KERNEL variable is set to 2.4.21 , e.g.

#!/bin/sh
if uname -r | grep -q ^2.6; then
  export LD_ASSUME_KERNEL=2.4.21
fi
exec /your/path/icm -p/your/path/ $*

How to calculate an ellipsoid approximating a grob or a set points?

For grobs, use this macro if you need only to know the lengths. macro calcEllips XYZ b=Eigen(Tensor(XYZ)) len1 = Length(b[?,1]) # the longest half-axis len2 = Length(b[?,2]) # the second half-axis len3 = Length(b[?,3]) # the third half-axis keep len1, len2, len3 endmacro XYZ is a coordinate matrix which can be returned by the Xyz function:

• Xyz( g_ ) # for grobs
• Xyz( as_ ) # for atom selections

calcEllips Xyz( a_/* )
show len1,len2,len3

For more detailed output in which the vectors themselves are returned, use the calcEllipsoid macro.

How to calculate residue properties of a certain group of residues, e.g. around a pocket?

Here is the recipe for arbitrary residue-based properties, which you define yourself.

• Find a selection of atoms or residues for which you want to find the property distribution, e.g. -- sph_atoms = Sphere( g_pocket1 a_rec. 2.5 ) -- # returns atoms of around the grob.
• Transform the selection in the residue level, e.g. -- sph_res = Res( sph_atoms ) --
• Turn that into a sequence, e.g. -- seq_sph = Sequence( sph_res ) --
• Create an array with the residue type values for all 26 letters of the alphabet, e.g. h={1.8,0.,2.5,-3.5,-3.5,2.8,-.4,-3.2,4.5,0.,-3.9,3.8,1.9,-3.5,.0,-1.6,-3.5,-4.5,-.8,-.7,0.,4.2,-.9,0.,-1.3,0.} for a hydrophobicity scale, (A, 1.8, is hydrophobic, and Y, -1.3, is not).
• Use the Rarray( seq_ , R_26 ) function to transfer the property, e.g. -- Rarray( seq_sph , h ) -- This function will return an array of values for each residue in the sequence.
• Use the Mean ,or Min ,or Max, or Rmsd functions to find the distribution properties, e.g. -- Mean( Rarray( seq_sph, h ) --

# user defined residue scale defined for A,B,C,D,....X,Y,Z
h={1.8,0.,2.5,-3.5,-3.5,2.8,-.4,-3.2,4.5,0.,-3.9,3.8,1.9,-3.5,.0,-1.6,-3.5,-4.5,-.8,-.7,0.,4.2,-.9,0.,-1.3,0.}

Mean(Rarray(Sequence(Res(Sphere(g_pocket1,a_rec.,2.5)))))