Using the system class

All of the tools in pele are modular, and can be used without a system class, but seting up a system class can make your life a lot easier. If you set it up once then access to all the tools in pele is simply one function call away. This tutorial is designed to show how use the system class to access the most common features of pele.


This tutorial mirrors an example in the examples folder.

This tutorial uses the LJCluster system class as an example. We start by initializing the system class. Different system classes may take different parameters for the constructor. LJCluster simply takes the number of atoms. We’ll look at clusters of 13 atoms:

from import LJCluster
natoms = 13
system = LJCluster(natoms)

We start by generating a random configuration:

coords = system.get_random_configuration()

We want to know the energy of that configuration, so we use the system class to create a potential object. This potential object has functions for computing the energy, gradient, and Hessian:

potential = system.get_potential()
energy = potential.getEnergy(coords)
print "the energy of the random configuration is", energy

Next we minimize that configuration to it’s nearest local minimum. To do that we use the system class to generate an object to do the minimization for us:

quencher = system.get_minimizer()
ret = quencher(coords)
newcoords = ret.coords
newenergy =
print "after quenching, the energy is", newenergy

Next we will create a database to store the newly generated minimum:

db = system.create_database()
minimum1 = db.addMinimum(newenergy, newcoords)


This creates a database in memory. If you want to save the results to a file you would use:

db = system.create_database("lj.sqlite")

Let’s generate a second random minimum and add it to the database. Now we use a shortcut from the system class:

ret = system.get_random_minimized_configuration()
print "a second minimum has energy", ret[1]
e2, coords2 = ret[1], ret[0]
minimum2 = db.addMinimum(e2, coords2)

Now we’ll do a short basinhopping run to find the global minimum and build up the database of minima:

bh = system.get_basinhopping(database=db)
niter = 20
print "the lowest energy found after", niter, " basinhopping steps is", db.minima()[0].energy

Print the energies of all the minima we’ve found:

print "the minima in the database have energies"
for minimum in db.minima():
    print "  ",

Next, lets find the minimum distance (a.k.a. mindist) between the two lowest minima:

m1, m2 = db.minima()[:2]
mindist = system.get_mindist()
dist, coords1, coords2 = mindist(m1.coords, m2.coords)
print "the minimum distance between the two lowest minima is", dist

Now lets do a double ended connect run. This finds a connected series of minima and transition states between two minima:

connect = system.get_double_ended_connect(m1, m2, db)
mints, S, energies = connect.returnPath()
nts = (len(mints) - 1)/2
print "found a connection with", nts, "transition states"

Finally, lets connect all of the minima in the database to the lowest minimum:

print "now connecting all the minima to the lowest energy minimum"
from pele.landscape import ConnectManager
manager = ConnectManager(db, strategy="gmin")
for i in xrange(db.number_of_minima()-1):
    print "connecting minima with id's", m1._id, m2._id
    m1, m2 = manager.get_connect_job()
    connect = system.get_double_ended_connect(m1, m2, db)

And we’ll end by printing out some information about what is in the database:

print "database summary:"
print "    ", len(db.minima()), "minima"
print "    ", len(db.transition_states()), "transition states"

Previous topic

Creating Your Potential

Next topic

Global Optimization