clustering Package

clustering Module

class AdaptivePELE.clustering.clustering.AltStructures

Bases: object

Helper class, each cluster will have an instance of AltStructures that will maintain a priority queue (pq) of alternative structures to spawn from encoded as tuples (priority, PDB).

addStructure(PDB, threshold, resname, resnum, resChain, contactThreshold, similarityEvaluator, trajPosition)

Perform a subclustering, with sub-clusters of size threshold/2

Parameters

• PDB (PDB) – Structure to cluster

• threshold (float) – Size of the cluster

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• contactThreshold (float) – Distance at which to atoms are considered in contact

• similarityEvaluator (SimilarityEvaluator) – Object that determinates the similarity between two structures

• trajPosition (int, int, int) – Tuple of (epoch, trajectory, snapshot) that permit identifying the structure added

altSpawnSelection(centerPair)

Select an alternative PDB from the cluster center to spawn from

Parameters

centerPair (int, PDB) – Tuple with the population of the representative structure and the PDB of said structure

Returns

PDB, tuple – PDB of the strucutre selected to spawn and tuple consisting of (epoch, trajectory, snapshot)

cleanPQ()

Ensure that the alternative structures priority queue has no more elements than the limit in order to ensure efficiency

sizePQ()

Get the number of sub-clusters stored in the priority queue

Returns

int – Number of sub-clusters stored in the priority queue

updateIndex()

Update the index which represents chronological order of entries in the priority queue

Returns

int – Index of the following element

class AdaptivePELE.clustering.clustering.CMClusteringEvaluator(similarityEvaluator, symmetryEvaluator)

Bases: AdaptivePELE.clustering.clustering.ClusteringEvaluator

Helper object to carry out the RMSD clustering

Parameters

• similarityEvaluator (CMSimilarityEvaluator) – object that calculates the similarity between two contact maps

• symmetryEvaluator (SymmetryContactMapEvaluator) – object to introduce the symmetry in the contacts maps

checkAttributes(pdb, resname, resnum, resChain, contactThresholdDistance)

Check wether all attributes are set for this iteration

Parameters

• pdb (PDB) – Structure to compare

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• contactThreshold (float) – Distance between two atoms to be considered in contact (default 8)

getInnerLimit(cluster)

Return the threshold of the cluster

Parameters

cluster (Cluster) – Cluster to compare

Returns

float – Threshold of the cluster

isElement(pdb, cluster, resname, resnum, resChain, contactThresholdDistance)

Evaluate wether a conformation is a member of a cluster

Parameters

• pdb (PDB) – Structure to compare

• cluster (Cluster) – Cluster to compare

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• contactThreshold (float) – Distance between two atoms to be considered in contact (default 8)

Returns

bool, float – Whether the structure belong to the cluster and the distance between them

limitMax = {4: 0.2, 6: 0.8, 8: 2, 10: 4}

limitSlope = {4: 60, 6: 15, 8: 6, 10: 3}

class AdaptivePELE.clustering.clustering.CMSimilarityEvaluator(typeEvaluator)

Bases: object

Evaluate the similarity of two contactMaps by calculating the ratio of the number of differences over the average of elements in the contacts maps, their correlation or their Jaccard index, that is, the ratio between the intersection of the two contact maps and their union

isSimilarCluster(contactMap, clusterContactMap, symContactMapEvaluator)

Evaluate if two contactMaps are similar or not, return True if yes, False otherwise

Parameters

• contactMap (numpy.Array) – contactMap of the structure to compare

• contactMap – contactMap of the structure to compare

• symContactMapEvaluator (SymmetryContactMapEvaluator) – Contact Map symmetry evaluator object

Returns

float – distance between contact maps

class AdaptivePELE.clustering.clustering.Cluster(pdb, thresholdRadius=None, contactMap=None, contacts=None, metrics=None, metricCol=None, density=None, contactThreshold=8, altSelection=False, trajPosition=None)

Bases: object

A cluster contains a representative structure(pdb), the number of elements, its density, threshold, number of contacts, a contactMap(sometimes) and a metric

Parameters

• pdb (PDB) – Pdb of the representative structure

• thresholdRadius (float) – Threshold of the cluster

• contactMap (numpy.Array) – The contact map of the ligand and the protein

• contacts (float) – Ratio of the number of alpha carbons in contact with the ligand

• metrics (numpy.Array) – Array of the metrics corresponding to the cluster

• metricCol (int) – Column of the prefered metric

• density (float) – Density of the cluster

• contactThreshold (float) – Distance between two atoms to be considered in contact (default 8)

• altSelection (bool) – Flag that controls wether to use the alternative structures (default 8)

• trajPosition (int, int, int) – Tuple of (epoch, trajectory, snapshot) that permit identifying the structure added

addElement(metrics)

Add a new element to the cluster

Parameters

metrics (numpy.Array) – Array of metrics of the new structure

getContacts()

Get the contacts ratio of the cluster

Returns

float – contact ratio of the cluster

getMetric()

Get the value of the prefered metric if present, otherwise return None

Returns

float – Value of the prefered metric

getMetricFromColumn(numcol)

Get the value of the metric in column numcol if present, otherwise return None

Parameters

numcol (int) – Column of the desired metric

Returns

float – Value of the prefered metric

printCluster(verbose=False)

Print cluster information

Parameters

verbose (bool) – Flag to control the verbosity of the code (default is False)

writePDB(path)

Write the pdb of the representative structure to file

Parameters

path (str) – Filename of the file to write

writeSpawningStructure(path)

Write the pdb of the chosen structure to spawn

Parameters

path (str) – Filename of the file to write

Returns int, int, int

Tuple of (epoch, trajectory, snapshot) that permit identifying the structure added

class AdaptivePELE.clustering.clustering.Clustering(resname='', resnum=0, resChain='', reportBaseFilename=None, columnOfReportFile=None, contactThresholdDistance=8, altSelection=False)

Bases: object

Base class for clustering methods, it defines a cluster method that contacts and accumulative inherit and use

Parameters

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

• contactThresholdDistance (float) – Distance at wich a ligand atom and a protein atom are considered in contact(default 8)

addSnapshotToCluster(trajNum, snapshot, origCluster, snapshotNum, metrics=None, col=None, topology=None)

Cluster a snapshot using the leader algorithm

Parameters

• trajNum (int) – Trajectory number

• snapshot (str) – Snapshot to add

• origCluster (int) – Cluster found in the previos snapshot

• snapshotNum (int) – Number of snapshot in its trajectory

• metrics (numpy.Array) – Array with the metrics of the snapshot

• col (int) – Column of the desired metrics

• topology (list) – Topology for non-pdb trajectories

Returns

int – Cluster to which the snapshot belongs

cluster(paths, ignoreFirstRow=False, topology=None, epoch=None, outputPathConstants=None)

Cluster the snaptshots contained in the paths folder

Parameters

• paths (list) – List of folders with the snapshots

• ignoreFirstRow (bool) – Flag wether to ignore the first snapshot of a trajectory

• topology (Topology) – Topology object containing the set of topologies needed for the simulation

• epoch (int) – Epoch number

• outputPathConstants (OutputPathConstants) – Contains outputPath-related constants

clusterIterator()

Iterator over the clusters

emptyClustering()

Delete previous results of clustering object

filterClustersAccordingToBox(simulationRunnerParams)

Filter the clusters to select only the ones whose representative structures will fit into the selected box

Parameters

simulationRunnerParams (SimulationParameters) – SimulationParameters Simulation parameters object

Returns list, list

– list of the filtered clusters, list of bools flagging wether the cluster is selected or not

filterClustersAccordingToMetric(clustersFiltered, filter_value, condition, col_filter)

Filter the clusters to select only the ones whose metric fits an specific criterion

Parameters

• clustersFiltered (list) – List of clusters to be processed

• filter_value (float) – Value to use in the filtering

• condition (str) – Whether to use > or < condition in the filtering

• col_filter (int) – Column of the report to use

Returns list, list

– list of the filtered clusters, list of bools flagging whether the cluster is selected or not

getCluster(clusterNum)

Get the cluster at index clusterNum

Returns

Cluster – Cluster at clusterNum

getClusterListForSpawning()

Return the clusters object to be used in the spawning

Returns

Clusters – Container object for the clusters

getMetricsFromColumn(col)

Get the metric of the clusters

Parameters

col (int) – Column to select the metric

Returns

np.array – Array containing the metric of the clusters

getNumberClusters()

Get the number of clusters

Returns

int – Number of clusters

getOptimalMetric(column=None, simulationType='min')

Find the cluster with the best metric

Parameters

• column (int) – Column of the metric that defines the best cluster, if not specified, the cluster metric is chosen

• simulationType (str) – Define optimal metric as the maximum or minimum, max or min

Returns

int – Number of cluster with the optimal metric

setCol(col)

Set the column of the prefered column to col

Parameters

col (int) – Column of the prefered column

setProcessors(processors)

updateRepeatParameters(repeat, steps)

Update parameters that should be extracted from the simulation object

Parameters

• repeat (bool) – Whether to avoid repeating steps (False for PELE, True for md)

• steps (int) – steps per epoch

writeClusterMetric(path, metricCol)

Write the metric of each node in the conformation network in a tab-separated file

Parameters

• path (str) – Path where to write the network

• metricCol (int) – Column of the metric of interest

writeConformationNodePopulation(path)

Write the population of each node in the conformation network in a tab-separated file

Parameters

path (str) – Path where to write the network

writeOutput(outputPath, degeneracy, outputObject, writeAll)

Writes all the clustering information in outputPath

Parameters

• outputPath (str) – Folder that will contain all the clustering information

• degeneracy (list) – Degeneracy of each cluster. It must be in the same order as in the self.clusters list

• outputObject (str) – Output name for the pickle object

• writeAll (bool) – Wether to write pdb files for all cluster in addition of the summary

writePathwayOptimalCluster(filename)

Extract the pathway to the cluster with the best metric as a trajectory and write it to a PDB file

Parameters

filename (str) – Path where to write the trajectory

writePathwayTrajectory(pathway, filename)

Write a list of cluster forming a pathway into a trajectory pdb file

Parameters

• pathway (list) – List of clusters that form the pathway

• filename (str) – Path where to write the trajectory

class AdaptivePELE.clustering.clustering.ClusteringBuilder

Bases: object

buildClustering(clusteringBlock, reportBaseFilename=None, columnOfReportFile=None)

Builder to create the appropiate clustering object

Parameters

• clusteringBlock (dict) – Parameters of the clustering process

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

Returns

Clustering – Clustering object selected

class AdaptivePELE.clustering.clustering.ClusteringEvaluator

Bases: object

cleanContactMap()

Clean the attributes to prepare for next iteration

class AdaptivePELE.clustering.clustering.Clusters

Bases: object

addCluster(cluster)

Add a new cluster

Parameters

cluster (Cluster) – Cluster object to insert

getCluster(clusterNum)

Get the cluster at position clusterNum

Parameters

clusterNum (int) – Index of the cluster to retrieve

Returns

Cluster – Cluster at position clusterNum

getNumberClusters()

Get the number of clusters contained

Returns

int – Number of clusters contained

insertCluster(index, cluster)

Insert a cluster in a specified index

Parameters

• index (int) – Positions at which insert the cluster

• cluster (Cluster) – Cluster object to insert

printClusters(verbose=False)

Print clusters information

Parameters

verbose (bool) – Flag to control the verbosity of the code (default is False)

class AdaptivePELE.clustering.clustering.ConformationNetwork

Bases: object

Object that contains the conformation network, a network with clusters as nodes and edges representing trantions between clusters. The network is stored using the networkx package[1]

1

Networkx python package https://networkx.github.io

add_edge(source, target)

Add an edge to the network (wrapper for networkx method)

Parameters

• source (int) – Name of the source node

• target (int) – Name of the target node

add_node(node, **kwargs)

Add a node to the network (wrapper for networkx method)

Parameters

• node (int) – Name of the node

• kwargs (keyword arguments, optional) – Set or change attributes using key=value.

createPathwayToCluster(clusterLeave)

Retrace the FDT from a specific cluster to the root where it was discovered

Parameters

clusterLeave (int) – End point of the pathway to reconstruct

Returns

list – List of snapshots conforming a pathway

writeConformationNetwork(path)

Write the conformational network to file to visualize it

Parameters

path (str) – Path where to write the network

writeFDT(path)

Write the first discovery tree to file in edgelist format to visualize it

Parameters

path (str) – Path where to write the network

class AdaptivePELE.clustering.clustering.ContactMapAccumulativeClustering(thresholdCalculator, similarityEvaluator, resname='', resnum=0, resChain='', reportBaseFilename=None, columnOfReportFile=None, contactThresholdDistance=8, symmetries=None, altSelection=False)

Bases: AdaptivePELE.clustering.clustering.Clustering

Cluster together all snapshots that have similar enough contactMaps. This similarity can be calculated with different methods (see similariyEvaluator documentation)

Parameters

• thresholdCalculator (ThresholdCalculator) – ThresholdCalculator object that calculate the threshold according to the contacts ratio

• similarityEvaluator (object) – object that calculates the similarity between two contact maps

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

• contactThresholdDistance (float) – Distance at wich a ligand atom and a protein atom are considered in contact(default 8)

• symmetries (list) – List of symmetric groups

• altSelection (bool) – Flag that controls wether to use the alternative structures (default 8)

class AdaptivePELE.clustering.clustering.ContactsClustering(thresholdCalculator, resname='', resnum=0, resChain='', reportBaseFilename=None, columnOfReportFile=None, contactThresholdDistance=8, symmetries=None, altSelection=False, useContacts=True)

Bases: AdaptivePELE.clustering.clustering.Clustering

Cluster together all snapshots that are closer to the cluster center than certain threshold. This threshold is assigned according to the ratio of number of contacts over the number of heavy atoms of the ligand

Parameters

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• thresholdCalculator (ThresholdCalculator) – ThresholdCalculator object that calculate the threshold according to the contacts ratio

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

• contactThresholdDistance (float) – Distance at wich a ligand atom and a protein atom are considered in contact(default 8)

• symmetries (list) – List of symmetric groups

• altSelection (bool) – Flag that controls wether to use the alternative structures (default 8)

• useContacts (bool) – Flag that controls whether to count the protein ligand contacts (useful mostly for ligand only simulations)

class AdaptivePELE.clustering.clustering.ContactsClusteringEvaluator(RMSDCalculator_object)

Bases: AdaptivePELE.clustering.clustering.ClusteringEvaluator

Helper object to carry out the RMSD clustering

Parameters

RMSDCalculator (RMSDCalculator) – object that calculates the RMSD between two conformations

checkAttributes(pdb, resname, resnum, resChain, contactThresholdDistance)

Check wether all attributes are set for this iteration

Parameters

• pdb (PDB) – Structure to compare

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• contactThreshold (float) – Distance between two atoms to be considered in contact (default 8)

getInnerLimit(cluster)

Return the threshold of the cluster

Parameters

cluster (Cluster) – Cluster to compare

Returns

float – Threshold of the cluster

isElement(pdb, cluster, resname, resnum, resChain, contactThresholdDistance)

Evaluate wether a conformation is a member of a cluster

Parameters

• pdb (PDB) – Structure to compare

• cluster (Cluster) – Cluster to compare

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• contactThreshold (float) – Distance between two atoms to be considered in contact (default 8)

Returns

bool, float – Whether the structure belong to the cluster and the distance between them

class AdaptivePELE.clustering.clustering.MSMClustering(n_clusters, tica=False, resname='', resnum=0, resChain='', symmetries=None, atom_Ids='', writeCA=False, sidechains=False, tica_lagtime=10, tica_nICs=3, tica_kinetic_map=True, tica_commute_map=False)

Bases: AdaptivePELE.clustering.clustering.Clustering

Cluster the trajectories to estimate a Markov State Model (MSM)

Base class for clustering methods, it defines a cluster method that contacts and accumulative inherit and use

Parameters

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

• contactThresholdDistance (float) – Distance at wich a ligand atom and a protein atom are considered in contact(default 8)

cluster(paths, topology=None, epoch=None, outputPathConstants=None)

Cluster the snaptshots contained in the paths folder

Parameters

• paths (list) – List of folders with the snapshots

• topology (Topology) – Topology object containing the set of topologies needed for the simulation

• epoch (int) – Epoch number

• outputPathConstants (OutputPathConstants) – Contains outputPath-related constants

filterClustersAccordingToBox(simulationRunnerParams)

Filter the clusters to select only the ones whose representative structures will fit into the selected box

Parameters

simulationRunnerParams (SimulationParameters) – SimulationParameters Simulation parameters object

Returns list, list

– list of the filtered clusters, list of bools flagging wether the cluster is selected or not

filterClustersAccordingToMetric(clustersFiltered, filter_value, condition, col_filter)

Filter the clusters to select only the ones whose metric fits an specific criterion

Parameters

• clustersFiltered (list) – List of clusters to be processed

• filter_value (float) – Value to use in the filtering

• condition (str) – Whether to use > or < condition in the filtering

• col_filter (int) – Column of the report to use

Returns list, list

– list of the filtered clusters, list of bools flagging whether the cluster is selected or not

getClusterListForSpawning()

Return the clusters object to be used in the spawning

Returns

Clusters – Container object for the clusters

setProcessors(processors)

updateRepeatParameters(repeat, steps)

Update parameters that should be extracted from the simulation object

Parameters

• repeat (bool) – Whether to avoid repeating steps (False for PELE, True for md)

• steps (int) – steps per epoch

writeOutput(outputPath, degeneracy, outputObject, writeAll)

Writes all the clustering information in outputPath

Parameters

• outputPath (str) – Folder that will contain all the clustering information

• degeneracy (list) – Degeneracy of each cluster. It must be in the same order as in the self.clusters list

• outputObject (str) – Output name for the pickle object

• writeAll (bool) – Wether to write pdb files for all cluster in addition of the summary

class AdaptivePELE.clustering.clustering.NullClustering

Bases: AdaptivePELE.clustering.clustering.Clustering

Don’t generate any clustering, works essentially as a placeholder for simulation when no clustering is desired

Base class for clustering methods, it defines a cluster method that contacts and accumulative inherit and use

Parameters

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

• contactThresholdDistance (float) – Distance at wich a ligand atom and a protein atom are considered in contact(default 8)

cluster(paths, topology=None, epoch=None, outputPathConstants=None)

Cluster the snaptshots contained in the paths folder

Parameters

• paths (list) – List of folders with the snapshots

• topology (Topology) – Topology object containing the set of topologies needed for the simulation

• epoch (int) – Epoch number

• outputPathConstants (OutputPathConstants) – Contains outputPath-related constants

writeOutput(outputPath, degeneracy, outputObject, writeAll)

Writes all the clustering information in outputPath

Parameters

• outputPath (str) – Folder that will contain all the clustering information

• degeneracy (list) – Degeneracy of each cluster. It must be in the same order as in the self.clusters list

• outputObject (str) – Output name for the pickle object

• writeAll (bool) – Wether to write pdb files for all cluster in addition of the summary

class AdaptivePELE.clustering.clustering.RMSDOnlyClusteringEvaluator(RMSDCalculator_object)

Bases: AdaptivePELE.clustering.clustering.ContactsClusteringEvaluator

Helper object to carry out the RMSD clustering

Parameters

RMSDCalculator (RMSDCalculator) – object that calculates the RMSD between two conformations

checkAttributes(pdb, resname, resnum, resChain, contactThresholdDistance)

Check wether all attributes are set for this iteration

Parameters

• pdb (PDB) – Structure to compare

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• contactThreshold (float) – Distance between two atoms to be considered in contact (default 8)

class AdaptivePELE.clustering.clustering.SequentialLastSnapshotClustering(resname='', resnum=0, resChain='', reportBaseFilename=None, columnOfReportFile=None, contactThresholdDistance=8, altSelection=False)

Bases: AdaptivePELE.clustering.clustering.Clustering

Assigned the last snapshot of the trajectory to a cluster. Only useful for PELE sequential runs

Base class for clustering methods, it defines a cluster method that contacts and accumulative inherit and use

Parameters

• resname (str) – String containing the three letter name of the ligand in the pdb

• resnum (int) – Integer containing the residue number of the ligand in the pdb

• resChain (str) – String containing the chain name of the ligand in the pdb

• reportBaseFilename (str) – Name of the file that contains the metrics of the snapshots to cluster

• columnOfReportFile (int) – Column of the report file that contain the metric of interest

• contactThresholdDistance (float) – Distance at wich a ligand atom and a protein atom are considered in contact(default 8)

addSnapshotToCluster(snapshot, metrics=None, col=None, topology=None)

Cluster a snapshot using the leader algorithm

Parameters

• trajNum (int) – Trajectory number

• snapshot (str) – Snapshot to add

• metrics (numpy.Array) – Array with the metrics of the snapshot

• col (int) – Column of the desired metrics

• topology (list) – Topology for non-pdb trajectories

Returns

int – Cluster to which the snapshot belongs

cluster(paths, topology=None, epoch=None, outputPathConstants=None)

Cluster the snaptshots contained in the paths folder

Parameters

• paths (list) – List of folders with the snapshots

• topology (Topology) – Topology object containing the set of topologies needed for the simulation

• epoch (int) – Epoch number

• outputPathConstants (OutputPathConstants) – Contains outputPath-related constants

AdaptivePELE.clustering.clustering.filterRepeatedReports(metrics, column=2)

Filter the matrix containing the report information to avoid rejected steps

Parameters

• metrics (np.ndarray) – Contents of the report file

• column (int) – Column to check for repeats

Returns

np.ndarray – Contents of the report file filtered

AdaptivePELE.clustering.clustering.getAllTrajectories(paths)

Find all the trajectory files in the paths specified

Parameters

paths (str) – The path where to find the trajectories

Returns

list – A list with the names of all the trajectories in paths

AdaptivePELE.clustering.clustering.loadReportFile(reportFile)

Load a report file and filter it

Parameters

reportFile (str) – Name of the report file

Returns

np.ndarray – Contents of the report file

class AdaptivePELE.clustering.clustering.similarityEvaluatorBuilder

Bases: object

build(similarityEvaluatorType)

Builder to create the appropiate similarityEvaluator

Parameters

similarityEvaluatorType (str) – Type of similarityEvaluator chosen

Returns

SimilarityEvaluator – SimilarityEvaluator object selected

clusteringTypes Module

class AdaptivePELE.clustering.clusteringTypes.CLUSTERING_TYPES

Bases: object

MSMClustering = 4

contactMap = 1

lastSnapshot = 2

null = 3

rmsd = 0

class AdaptivePELE.clustering.clusteringTypes.SIMILARITY_TYPES

Bases: object

Jaccard = 1

correlation = 2

differenceDistance = 0

thresholdcalculator Module

class AdaptivePELE.clustering.thresholdcalculator.ThresholdCalculator

Bases: object

abstract calculate(contacts)

class AdaptivePELE.clustering.thresholdcalculator.ThresholdCalculatorBuilder

Bases: object

build(clusteringBlock)

Bulid the selecte thresholdCaulcualtor object

Parameters

clusteringBlock (dict) – Parameters block corresponding to the threshold calculator

Returns

ThresholdCalculator – thresholdCalculator object selected

class AdaptivePELE.clustering.thresholdcalculator.ThresholdCalculatorConstant(value=2)

Bases: AdaptivePELE.clustering.thresholdcalculator.ThresholdCalculator

calculate(contacts)

Calculate the threshold value of a cluster. In this case it is constant, the contacts ratio is only passed for compatibility purposes

Parameters

contacts (float) – Contact ratio

Returns

float – threshold value of the cluster

getMaxThreshold()

Method that returns the maximum treshold possible, required for new distance-ordered clustering(in early development)

Returns

float – Maximum threshold possible

class AdaptivePELE.clustering.thresholdcalculator.ThresholdCalculatorHeaviside(conditions=None, values=None)

Bases: AdaptivePELE.clustering.thresholdcalculator.ThresholdCalculator

calculate(contacts)

Calculate the threshold value of a cluster according to the contacts ratio and the selected conditions and values

Parameters

contacts (float) – Contact ratio

Returns

float – threshold value of the cluster

getMaxThreshold()

Method that returns the maximum treshold possible, required for new distance-ordered clustering(in early development)

Returns

float – Maximum threshold possible

thresholdcalculatortypes Module

class AdaptivePELE.clustering.thresholdcalculatortypes.THRESHOLD_CALCULATOR_TYPES

Bases: object

constant = 1

heaviside = 0