Added Example and fixed documentation

.github/workflows/CI.yaml

@@ -79,4 +79,4 @@ jobs:
       shell: bash
       if: runner.os == 'Linux'
       run: |
-        CodeEntropyPoseidon -f "CodeEntropy/tests/data/1AKI_prod_60.tpr" "CodeEntropy/tests/data/1AKI_prod_60.trr" -l "protein" -b 2 -e 12 -d 2 -t 2 --wm --res --uatom --topog 5 --solwm --solres --soluatom --solContact
+        CodeEntropyPoseidon -f "CodeEntropy/tests/data/1AKI_prod_60.tpr" "CodeEntropy/tests/data/1AKI_prod_60.trr" -l "protein" -b 2 -e 12 -d 2 -t 2 --wm --res --uatom --topog 3 --solwm --solres --soluatom --solContact

.github/workflows/publish-on-pypi.yaml

@@ -0,0 +1,21 @@
+name: Publish on PyPI
+
+on:
+  push:
+    tags:
+      # After vMajor.Minor.Patch _anything_ is allowed (without "/") !
+      - '*'
+
+jobs:
+  publish:
+    name: Publish to pypi if tagged
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v2
+      - uses: actions/setup-python@v2
+      - uses: casperdcl/deploy-pypi@v2
+        with:
+          password: ${{ secrets.PYPI_API_TOKEN }}
+          pip: wheel -w dist/ --no-deps .
+          # only upload if a tag is pushed (otherwise just build & check)
+          upload: ${{ github.event_name == 'push' && startsWith(github.event.ref, 'refs/tags') }} 

CodeEntropy/ClassCollection/PoseidonClass.py

@@ -5,6 +5,8 @@
 from collections import Counter
 from datetime import datetime
 
+from joblib import delayed
+
 from CodeEntropy.poseidon.extractData.readFiles import populateTopology, getCoordsForces, getDistArray
 # # Energy is not needed
 # from CodeEntropy.poseidon.extractData.readFiles import populateEnergy, UAEnergyGroup
@@ -19,9 +21,15 @@
 from CodeEntropy.poseidon.analysis.EECalculation import processEE
 from CodeEntropy.poseidon.analysis.helper import memoryInfo, weightingPopulation
 
+from CodeEntropy.FunctionCollection.PoseidonHelper import frame_iteration
+
 from functools import partial
 from multiprocessing import Pool
 
+from pathos.pools import ProcessPool
+
+import numpy as nmp
+
 class Poseidon():
 	"""
 	Container to host data from MDAnalysis.Universe and run Poseidon Analysis
@@ -413,21 +421,37 @@ def __init__(self ,container, start=0, end=-1,
 		step=1, pureAtomNum=1, cutShell=None, 
 		excludedResnames=None,
 		water='WAT', verbose=False, thread=4):
-		"""This is a initialization function to collect information from a MDanalysis universe into a data container for analysis using POSEIDON
-		Pending Work
-			- rewrite this into a class initalization for easier understanding
-			- rewrite the energy part to be compulsory
-
-		Args:
-			container (MDAnalysis.Universe): A MDAnalysis object with coordinates, forces and energy (loaded to velocity field with the value [Potential Energy(index 0), Kinetic Energy(1) and dummy(2)] respectively)
-			start (int): Starting Frame ID. Defaults to 0.
-			end (int): Ending Frame ID, this frame is not included. Defaults to -1.
-			step (int, optional): Steps between frame. Defaults to 1.
-			pureAtomNum (int, optional): Reference molecule resid for pure liquid. Defaults to 1.
-			cutShell (float, optional): Explicit cut off shell (might be buggy since part of it is not defined). Default to None which uses the relative angular distance (RAD) algorithm. See Jonathan Higham and Richard H. Henchman , "Locally adaptive method to define coordination shell", J. Chem. Phys. 145, 084108 (2016)
-			excludedResnames (list, optional): List of resnames to exclude from nearest non alike analysis. Defaults to None.
-			water (str, optional): Resname for water molecules. Defaults to 'WAT'.
-			verbose (bool, optional): print out progress of each analysis step. Defaults to False.
+		"""
+
+		This is a initialization function to collect information from a MDanalysis universe into a data container for analysis using POSEIDON.
+
+		.. Warning ::
+
+			This requires a lot of memory to run with very little performance gain.
+
+		Parameters
+		----------
+		container : MDAnalyse.Universe
+			A Universe object will all topology, dihedrals,coordinates and force information Check ``Example/create_new_universe.py`` on how to create a universe from existing data.
+		start : int or None, Optional, default: 0
+			Frame id to start analysis.
+		end : int or None, Optional, default: -1
+			Frame id to end analysis.
+		step : int, Optional, default: 1
+			Steps between frame.
+		pureAtomNum : int, Optional, default: 1
+			Reference molecule resid for pure liquid.
+		cutShell : float or None, Optional, default: 1
+			Explicit cut off shell. Default to None which uses the relative angular distance (RAD) algorithm. See Jonathan Higham and Richard H. Henchman , "Locally adaptive method to define coordination shell", J. Chem. Phys. 145, 084108 (2016)
+		excludedResnames : list of str or None, Optional, default: None
+			List of resnames to exclude from nearest non alike analysis.
+		water : list of str, Optional, default: "WAT"
+			Resname for water molecules.
+		verbose : bool, Optional, default: False
+			Print out progress of each step.
+		thread : int, Optional, default: 4
+			How many multiprocess thread to spawn
+
 		"""
 		startTime = datetime.now()
 		print(startTime)
@@ -463,9 +487,15 @@ def __init__(self ,container, start=0, end=-1,
 		if end == -1:
 			end = len(container.trajectory)
 		sys.setrecursionlimit(3000000)
-		p = Pool(processes=thread)
-		fi_partial = partial(self.frame_iteration, container, all_data, dimensions, startTime, verbosePrint, waterTuple, cutShell, excludedResnames)
-		data = p.map(fi_partial, [frame for frame in range(int(start), int(end), int(step))])
+		#p = Pool(processes=thread)
+		p = ProcessPool(nodes=thread)
+		fi_partial = partial(frame_iteration, container, all_data, dimensions, startTime, verbosePrint, waterTuple, cutShell, excludedResnames)
+		range = nmp.arange(int(start), int(end), int(step))
+		data = p.map(fi_partial, range)
+		p.close()
+
+		print("Done Threading Work")
+		print(len(data))
 		# data = []
 		# for frame in range(int(start), int(end), int(step)):
 		#     data.append(fi_partial(frame))
@@ -480,120 +510,3 @@ def __init__(self ,container, start=0, end=-1,
 		print(datetime.now() - startTime)
 
 		self.allMoleculeList = allMoleculeList
-
-	def frame_iteration(self, container, all_data, dimensions, startTime, verbosePrint, waterTuple, cutShell, excludedResnames, frame):
-		clearClass(all_data)
-		print(f"frame = {frame}")
-
-		all_data, dimensions = getCoordsForces(container, 
-				all_data, dimensions, frame, startTime, 
-				verbosePrint)
-		# # Energy is not needed
-		# populateEnergy(container, all_data, 
-		#         dimensions, frame, startTime, verbosePrint)
-		# UAEnergyGroup(all_data)
-
-
-		calculateDihedrals(all_data, dimensions)
-		verbosePrint('DIH')
-		verbosePrint(datetime.now() - startTime)
-		sys.stdout.flush()
-
-		#'''
-		traj = container.trajectory[frame]
-		neighbour_coords = None
-		neighbour_coords = traj.positions
-
-		max_cutoff = 10
-		for x in range(0, len(all_data)):
-			atom = all_data[x]
-			#start nearest array from solutes
-			if atom.resname not in waterTuple:
-				getDistArray(atom, all_data, traj, max_cutoff,
-						dimensions, neighbour_coords,
-						startTime, verbosePrint)
-				#find nearest array for solute neighbours that are solvent
-				for nearDist in atom.nearest_all_atom_array[0:20]:
-					neighbour = all_data[nearDist[0]]
-					if neighbour.resname in waterTuple and \
-							neighbour.nearest_all_atom_array == None:
-						getDistArray(neighbour, all_data, traj, max_cutoff,
-								dimensions, neighbour_coords,
-								startTime, verbosePrint)
-					#find solvent neighbours neighbours nearest array
-					if neighbour.nearest_all_atom_array != None:
-						for nearDist2 in neighbour.nearest_all_atom_array[0:20]:
-							neighbour2 = all_data[nearDist2[0]]
-							if neighbour2.resname in waterTuple and \
-									neighbour2.nearest_all_atom_array == None:
-								getDistArray(neighbour2, all_data, 
-										traj, max_cutoff,
-										dimensions, 
-										neighbour_coords,
-										startTime, verbosePrint)
-							else:
-								continue
-					else:
-						continue
-			else:
-				continue
-		verbosePrint('NEAREST ARRAYS')
-		verbosePrint(datetime.now() - startTime)
-		sys.stdout.flush()
-		#'''
-
-
-
-		if cutShell != None:
-			#Used for fixed cut-off coordination shells
-			try:
-				cutoff_dist = float(cutShell)
-				distCutoffNc(all_data, dimensions, cutoff_dist)
-				# #don't know what this is it is not defined anywhere
-				# NcPairs(all_data, dimensions)
-				verbosePrint('distCutoffNc')
-				verbosePrint(datetime.now() - startTime)
-				sys.stdout.flush() 
-			except ValueError:
-				logging.error('Cutoff distance needs to be a float')
-
-
-		if cutShell == None:
-			UALevelRAD(all_data, dimensions)
-			verbosePrint('RAD')
-			verbosePrint(datetime.now() - startTime)
-
-
-		#if inputType != 'pdb':
-		HBCalc(all_data, waterTuple, dimensions)
-		verbosePrint('HB')
-		verbosePrint(datetime.now() - startTime)
-		sys.stdout.flush() 
-
-
-		getShellAssignment(all_data, excludedResnames, dimensions, 
-				startTime, verbosePrint)
-		verbosePrint('PROX')
-		verbosePrint(datetime.now() - startTime)
-		sys.stdout.flush()
-
-
-		#if (force != None):
-		calculateFTMatrix(all_data, dimensions)
-		verbosePrint('FTMATRIX')
-		verbosePrint(datetime.now() - startTime)
-		sys.stdout.flush()
-
-
-		moleculePositionRankingRAD(all_data, waterTuple, dimensions)
-		verbosePrint('ORIENTS')
-		verbosePrint(datetime.now() - startTime)
-		sys.stdout.flush()
-
-		'''
-		if iterations == 1:
-			verbosePrint('Generating .pdb file for frame %s' % (frame+1))
-			pdbGenerate(all_data, ('frame_%s' % (frame+1)), dimensions)
-		'''
-
-		return (all_data, frame, dimensions)

CodeEntropy/FunctionCollection/PoseidonHelper.py

@@ -0,0 +1,135 @@
+import sys
+import logging
+
+from CodeEntropy.poseidon.extractData.readFiles import populateTopology, getCoordsForces, getDistArray
+# # Energy is not needed
+# from CodeEntropy.poseidon.extractData.readFiles import populateEnergy, UAEnergyGroup
+from CodeEntropy.poseidon.extractData.HBRAD import distCutoffNc, UALevelRAD, HBCalc
+from CodeEntropy.poseidon.extractData.forceTorques import calculateFTMatrix
+from CodeEntropy.poseidon.extractData.dihedrals import calculateDihedrals
+from CodeEntropy.poseidon.extractData.nearestNonlike2 import getShellAssignment, moleculePositionRankingRAD 
+from CodeEntropy.poseidon.extractData.outputFiles import moleculeObjectPopulation
+from CodeEntropy.poseidon.extractData.mainClass import clearClass
+
+from CodeEntropy.poseidon.analysis.populateClasses import classPopulation
+from CodeEntropy.poseidon.analysis.EECalculation import processEE
+from CodeEntropy.poseidon.analysis.helper import memoryInfo, weightingPopulation
+
+from datetime import datetime
+
+def frame_iteration(container, all_data, dimensions, startTime, verbosePrint, waterTuple, cutShell, excludedResnames, frame):
+    clearClass(all_data)
+    print(f"frame = {frame}")
+
+    all_data, dimensions = getCoordsForces(container, 
+            all_data, dimensions, frame, startTime, 
+            verbosePrint)
+    # # Energy is not needed
+    # populateEnergy(container, all_data, 
+    #         dimensions, frame, startTime, verbosePrint)
+    # UAEnergyGroup(all_data)
+
+
+    calculateDihedrals(all_data, dimensions)
+    verbosePrint('DIH')
+    verbosePrint(datetime.now() - startTime)
+    sys.stdout.flush()
+
+    #'''
+    traj = container.trajectory[frame]
+    neighbour_coords = None
+    neighbour_coords = traj.positions
+
+    max_cutoff = 10
+    for x in range(0, len(all_data)):
+        atom = all_data[x]
+        #start nearest array from solutes
+        if atom.resname not in waterTuple:
+            getDistArray(atom, all_data, traj, max_cutoff,
+                    dimensions, neighbour_coords,
+                    startTime, verbosePrint)
+            #find nearest array for solute neighbours that are solvent
+            for nearDist in atom.nearest_all_atom_array[0:20]:
+                neighbour = all_data[nearDist[0]]
+                if neighbour.resname in waterTuple and \
+                        neighbour.nearest_all_atom_array == None:
+                    getDistArray(neighbour, all_data, traj, max_cutoff,
+                            dimensions, neighbour_coords,
+                            startTime, verbosePrint)
+                #find solvent neighbours neighbours nearest array
+                if neighbour.nearest_all_atom_array != None:
+                    for nearDist2 in neighbour.nearest_all_atom_array[0:20]:
+                        neighbour2 = all_data[nearDist2[0]]
+                        if neighbour2.resname in waterTuple and \
+                                neighbour2.nearest_all_atom_array == None:
+                            getDistArray(neighbour2, all_data, 
+                                    traj, max_cutoff,
+                                    dimensions, 
+                                    neighbour_coords,
+                                    startTime, verbosePrint)
+                        else:
+                            continue
+                else:
+                    continue
+        else:
+            continue
+    verbosePrint('NEAREST ARRAYS')
+    verbosePrint(datetime.now() - startTime)
+    sys.stdout.flush()
+    #'''
+
+
+
+    if cutShell != None:
+        #Used for fixed cut-off coordination shells
+        try:
+            cutoff_dist = float(cutShell)
+            distCutoffNc(all_data, dimensions, cutoff_dist)
+            # #don't know what this is it is not defined anywhere
+            # NcPairs(all_data, dimensions)
+            verbosePrint('distCutoffNc')
+            verbosePrint(datetime.now() - startTime)
+            sys.stdout.flush() 
+        except ValueError:
+            logging.error('Cutoff distance needs to be a float')
+
+
+    if cutShell == None:
+        UALevelRAD(all_data, dimensions)
+        verbosePrint('RAD')
+        verbosePrint(datetime.now() - startTime)
+
+
+    #if inputType != 'pdb':
+    HBCalc(all_data, waterTuple, dimensions)
+    verbosePrint('HB')
+    verbosePrint(datetime.now() - startTime)
+    sys.stdout.flush() 
+
+
+    getShellAssignment(all_data, excludedResnames, dimensions, 
+            startTime, verbosePrint)
+    verbosePrint('PROX')
+    verbosePrint(datetime.now() - startTime)
+    sys.stdout.flush()
+
+
+    #if (force != None):
+    calculateFTMatrix(all_data, dimensions)
+    verbosePrint('FTMATRIX')
+    verbosePrint(datetime.now() - startTime)
+    sys.stdout.flush()
+
+
+    moleculePositionRankingRAD(all_data, waterTuple, dimensions)
+    verbosePrint('ORIENTS')
+    verbosePrint(datetime.now() - startTime)
+    sys.stdout.flush()
+
+    '''
+    if iterations == 1:
+        verbosePrint('Generating .pdb file for frame %s' % (frame+1))
+        pdbGenerate(all_data, ('frame_%s' % (frame+1)), dimensions)
+    '''
+    #print(all_data, frame, dimensions)
+    return (all_data, frame, dimensions)