From: Lu Yu
Date: 7 November 2011 16:31
Hi,
I was trying to use SHELXD to solve peptide structure. But I got stuck in the input .ins file, and I need some advice.
In the .ins file,
TITLE
CELL
ZERR
LATT
SYMM
SFAC C H N O
UNIT
FIND
PLOP
NTRY
HKLF
END
A rough estimate, there will be 62 C, 122 H, 14 N, 32 O in one unit cell.
1 for UNIT command, should the following nums be the num of atoms(C H N O) per unit cell and multiplied by 4? 62x4 for C, 122x4 for H, etc?
2 for FIND, the manual says 'estimated num of sites within 20% of true number', so should it be 20% of the total number of atoms in one unit cell? (62＋122＋14＋32)x 20%?
3 for PLOP, the manual says '# of peaks to start with in each cycle, Peaks are then eliminated one at a time until either the correlation coefficient cannot be increased any more or 50% of the peaks have been eliminated' & 'one should specify more than the expected number of atoms because this procedure involves the elimination of the 'wrong' atoms' which I don't fully understand, should the following nums be bigger than the total num of atoms in one unit cell?
Thanks in advance!
Lu
----------
From: Lu Yu
----------
From: George M. Sheldrick
UNIT specifies the number of atoms of each type in the unit-cell. For such
'small-molecule' problems you should try to get the numbers of heavier atoms
correct, if only CHNO are present any numbers will do.
For such problems I recommend setting FIND to about 70% of the number of atoms
(excluding H) in the asymmetric unit.
The first PLOP number should be approximately the number of atoms (excluding H)
in the asymmetric unit. The second PLOP number should be about 1.2 times this
and the third about 1.4 times it (three PLOP cycles are enough). This allows
the 'peaklist optimization algorithm' to throw out some of the atoms.
You will need data to 1.2A or better (1.0 is much better than 1.2!). The data
should be as complete as possible.
The following comments apply to larger problems, your structures is probably
small enough to ignore them (but you may still need a large NTRY).
I strongly recommend using the beta-test multiple-CPU version of shelxd. Direct
methods can be very computationally intensive for large structures. I recently
used it to solve a RNA with 1.0A data; on an 8-CPU machine it produced one
solution in a week. This version is faster even with one CPU, the standard
version would have taken about 3 months. I have seen several cases that had one
solution in 50000 or more trials (NTRY).
If you expect a partly helical structure or have a good partial model and have
data to 2.1A or better I strongly recommend ARCIMBOLDO. If all else fails I
would be happy to look at the data for you.
George
--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
Date: 7 November 2011 16:31
Hi,
I was trying to use SHELXD to solve peptide structure. But I got stuck in the input .ins file, and I need some advice.
In the .ins file,
TITLE
CELL
ZERR
LATT
SYMM
SFAC C H N O
UNIT
FIND
PLOP
NTRY
HKLF
END
A rough estimate, there will be 62 C, 122 H, 14 N, 32 O in one unit cell.
1 for UNIT command, should the following nums be the num of atoms(C H N O) per unit cell and multiplied by 4? 62x4 for C, 122x4 for H, etc?
2 for FIND, the manual says 'estimated num of sites within 20% of true number', so should it be 20% of the total number of atoms in one unit cell? (62＋122＋14＋32)x 20%?
3 for PLOP, the manual says '# of peaks to start with in each cycle, Peaks are then eliminated one at a time until either the correlation coefficient cannot be increased any more or 50% of the peaks have been eliminated' & 'one should specify more than the expected number of atoms because this procedure involves the elimination of the 'wrong' atoms' which I don't fully understand, should the following nums be bigger than the total num of atoms in one unit cell?
Thanks in advance!
Lu
----------
From: Lu Yu
----------
From: George M. Sheldrick
UNIT specifies the number of atoms of each type in the unit-cell. For such
'small-molecule' problems you should try to get the numbers of heavier atoms
correct, if only CHNO are present any numbers will do.
For such problems I recommend setting FIND to about 70% of the number of atoms
(excluding H) in the asymmetric unit.
The first PLOP number should be approximately the number of atoms (excluding H)
in the asymmetric unit. The second PLOP number should be about 1.2 times this
and the third about 1.4 times it (three PLOP cycles are enough). This allows
the 'peaklist optimization algorithm' to throw out some of the atoms.
You will need data to 1.2A or better (1.0 is much better than 1.2!). The data
should be as complete as possible.
The following comments apply to larger problems, your structures is probably
small enough to ignore them (but you may still need a large NTRY).
I strongly recommend using the beta-test multiple-CPU version of shelxd. Direct
methods can be very computationally intensive for large structures. I recently
used it to solve a RNA with 1.0A data; on an 8-CPU machine it produced one
solution in a week. This version is faster even with one CPU, the standard
version would have taken about 3 months. I have seen several cases that had one
solution in 50000 or more trials (NTRY).
If you expect a partly helical structure or have a good partial model and have
data to 2.1A or better I strongly recommend ARCIMBOLDO. If all else fails I
would be happy to look at the data for you.
George
--
Prof. George M. Sheldrick FRS
Dept. Structural Chemistry,
University of Goettingen,
Tammannstr. 4,
D37077 Goettingen, Germany
No comments:
Post a Comment