merge update

.github/workflows/main.yml

@@ -13,7 +13,7 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: ["2.7", "3.7", "3.8", "3.9", "3.10", "3.11"]
+        python-version: ["2.7", "3.8", "3.9", "3.10", "3.11"]
 
     steps:
     - uses: actions/checkout@v2
@@ -26,11 +26,11 @@ jobs:
         echo $CONDA/bin >> $GITHUB_PATH
     - name: Install dependencies
       run: |
-        if [[ ${{ matrix.python-version }} == "3.11" ]]; then conda config --add channels conda-forge; fi
+        if [[ ${{ matrix.python-version }} != "2.7" ]]; then conda config --add channels conda-forge; fi
         conda create --yes -n test python=${{ matrix.python-version }}
         source activate test
         conda install --yes numpy scipy nose pyparsing requests
-        if [[ ${{ matrix.python-version }} == "2.7" ]]; then conda install --yes unittest2; fi
+        if [[ ${{ matrix.python-version }} == "2.7" ]]; then conda install --yes unittest2; else conda install --yes pdbfixer; fi
         pip install mmtf-python
         pip install .
         python setup.py build_ext --inplace --force

README.rst

@@ -72,3 +72,19 @@ with ProDy. The original CE method was developed by Ilya Shindyalov and Philip
 Bourne. The Python version which is used by ProDy is developed by Jason Vertrees 
 and available under the New BSD license. 
 
+Hbp module: The calculation of hydrophobic interactions, solvent accessible surface 
+area (SASA) and volume for each residue is using geometric methods based on the 
+information of the atoms in the molecule. The methods have been programmed in C++ 
+and can be compiled as a python module “hpb.so” which is then used by ProDy.
+Files for compilation are stored at prody/proteins/hpbmodule folder and
+required C++ and Fortran compiler. After compilation hpb.so file can be
+stored in prody/proteins folder in ProDy or in the local directory which
+is used to perform calulations. The precompiled versions for Python 2.7,
+3.8, 3.9, and 3.10 are availabe in prody/proteins/hpbmodule. The user can
+choose the correct version of hpb.so and copy to the prody/proteins or
+local directory.
+C++ code of hpb.so was developed by Xin Cao and Fortran code by Xin Cao, 
+Michelle H. Hummel, Bihua Yu, and Evangelos A. Coutsias (License in 
+prody/proteins/hpbmodule folder). Details of the method can be found 
+in the Supplementary Material of InSty manuscript 
+(soon will be submitted for publication). 

docs/docs

@@ -1 +1 @@
-../docs
+ProDy/docs

docs/reference/proteins/interactions.rst

@@ -0,0 +1,6 @@
+Interactions Tools
+======================
+
+.. automodule:: prody.proteins.interactions
+   :members:
+   :undoc-members:

prody/apps/prody_apps/prody_select.py

@@ -35,7 +35,7 @@ def prody_select(selstr, *pdbs, **kwargs):
     prefix = kwargs.get('prefix', None)
     suffix = kwargs.get('suffix', '_selected')
     output = kwargs.get('output', None)
-    altloc = kwargs.get('altloc', None)
+    altloc = kwargs.get('altloc', 'all')
 
     for pdb in pdbs:
         pdb = parsePDB(pdb, altloc=altloc)
@@ -83,7 +83,7 @@ def addCommand(commands):
         type=str, help=('output filename prefix (default: PDB filename)'))
 
     group.add_argument('-L', '--altloc', dest='altloc', metavar='STR',
-        type=str, help=('altloc (default: None (take all))'))
+        type=str, default='all', help=('altloc (default: %(default)s)'))
 
     group.add_argument('-x', '--suffix', dest='suffix', metavar='STR',
         type=str, default='_selected',

prody/atomic/atom.py

@@ -94,6 +94,27 @@ def setCoords(self, coords):
         self._ag._coords[acsi, self._index] = coords
         self._ag._setTimeStamp(acsi)
 
+    def getAnisou(self):
+        """Returns a copy of anisotropic temperature factors of the atom from the active coordinate
+        set."""
+
+        if self._ag._anisous is not None:
+            return self._ag._anisous[self.getACSIndex(), self._index].copy()
+
+    def _getAnisou(self):
+        """Returns a view of anisotropic temperature factors of the atom from the active coordinate
+        set."""
+
+        if self._ag._anisous is not None:
+            return self._ag._anisous[self.getACSIndex(), self._index]
+
+    def setAnisou(self, anisou):
+        """Set anisotropic temperature factors of the atom in the active coordinate set."""
+
+        acsi = self.getACSIndex()
+        self._ag._anisous[acsi, self._index] = anisou
+        self._ag._setTimeStamp(acsi)
+
     def getCoordsets(self, indices=None):
         """Returns a copy of coordinate set(s) at given *indices*."""
 

prody/atomic/atomgroup.py

@@ -9,7 +9,8 @@
 
 from prody import LOGGER, PY2K
 from prody.kdtree import KDTree
-from prody.utilities import checkCoords, rangeString, getDistance, copy
+from prody.utilities import (checkCoords, checkAnisous, 
+                             rangeString, getDistance, copy)
 
 from .atomic import Atomic
 from .fields import ATOMIC_FIELDS, READONLY
@@ -130,7 +131,7 @@ class AtomGroup(Atomic):
                  '_donors', '_acceptors', '_nbexclusions', '_crossterms',
                  '_cslabels', '_acsi', '_n_csets', '_data',
                  '_fragments', '_flags', '_flagsts', '_subsets',
-                 '_msa', '_sequenceMap']
+                 '_msa', '_sequenceMap', '_anisous']
 
     def __init__(self, title='Unnamed'):
 
@@ -170,6 +171,7 @@ def __init__(self, title='Unnamed'):
         self._subsets = None
         self._msa = None
         self._sequenceMap = None
+        self._anisous = None
 
     def __repr__(self):
 
@@ -238,12 +240,20 @@ def __add__(self, other):
         if self._n_csets:
             if self._n_csets == other._n_csets:
                 new.setCoords(np.concatenate((self._coords, other._coords), 1))
+                this = self._anisous
+                that = other._anisous
+                if this is not None and that is not None:
+                    if (isinstance(this, np.ndarray) and isinstance(that, np.ndarray)
+                        and len(this) > 0 and len(that) > 0):
+                        new.setAnisous(np.concatenate((self._anisous, other._anisous), 1))
                 if self._n_csets > 1:
                     LOGGER.info('All {0} coordinate sets are copied to '
                                 '{1}.'.format(self._n_csets, new.getTitle()))
             else:
                 new.setCoords(np.concatenate((self._getCoords(),
                                               other._getCoords())))
+                new.setAnisous(np.concatenate((self.getAnisous(),
+                                               other.getAnisous())))
                 LOGGER.info('Active coordinate sets are copied to {0}.'
                             .format(new.getTitle()))
         elif other._n_csets:
@@ -565,7 +575,101 @@ def _setCoords(self, coords, label='', overwrite=False):
             self._setTimeStamp(acsi)
             self._cslabels[acsi] = str(label)
 
-    def addCoordset(self, coords, label=None):
+    def getAnisous(self):
+        """Returns a copy of anisotropic temperature factors from active coordinate set."""
+
+        if self._anisous is not None:
+            return self._anisous[self._acsi].copy()
+
+    def _getAnisous(self):
+        """Returns a view of anisotropic temperature factors from active coordinate set."""
+
+        if self._anisous is not None:
+            return self._anisous[self._acsi]
+
+    def setAnisous(self, anisous, label=''):
+        """Set anisotropic temperature factors of atoms. *anisous* may be any array like object
+        or an object instance with :meth:`getAnisous` method.  If the shape of
+        anisou array is ``(n_csets > 1, n_atoms, 3)``, it will replace all
+        coordinate sets and the active coordinate set index  will reset to
+        zero.  This situation can be avoided using :meth:`addCoordset`.
+        If shape of *coords* is ``(n_atoms, 3)`` or ``(1, n_atoms, 3)``, it
+        will replace the active coordinate set.  *label* argument may be used
+        to label coordinate set(s).  *label* may be a string or a list of
+        strings length equal to the number of coordinate sets."""
+
+        atoms = anisous
+        try:
+            if self._anisous is None and hasattr(atoms, '_getAnisous'):
+                anisous = atoms._getAnisous()
+            else:
+                anisous = atoms.getAnisous()
+        except AttributeError:
+            if self._anisous is None:
+                anisous = np.array(anisous)
+        else:
+            if anisous is None:
+                raise ValueError('anisous of {0} are not set'
+                                 .format(str(atoms)))
+
+        try:
+            checkAnisous(anisous, csets=True, dtype=(float, np.float32))
+        except TypeError:
+            raise TypeError('anisous must be a numpy array or an '
+                            'object with `getAnisous` method')
+
+        self._setAnisous(anisous, label=label)
+
+    def _setAnisous(self, anisous, label='', overwrite=False):
+        """Set anisotropic temperature factors without data type checking.
+        *anisous* must be a :class:`~numpy.ndarray`, but may have data type
+        other than :class:`~numpy.float64`, e.g. :class:`~numpy.float32`.
+        *label* argument may be used to label coordinate sets.  *label* may be
+        a string or a list of strings length equal to the number of
+        coordinate sets."""
+
+        n_atoms = self._n_atoms
+        if n_atoms:
+            if anisous.shape[-2] != n_atoms:
+                raise ValueError('anisous array has incorrect number of atoms')
+        else:
+            self._n_atoms = n_atoms = anisous.shape[-2]
+
+        ndim = anisous.ndim
+        shape = anisous.shape
+        if self._anisous is None or overwrite or (ndim == 6 and shape[0] > 1):
+            if ndim == 2:
+                self._anisous = anisous.reshape((1, n_atoms, 6))
+                self._cslabels = [str(label)]
+                self._n_csets = n_csets = 1
+
+            else:
+                self._anisous = anisous
+                self._n_csets = n_csets = shape[0]
+
+                if isinstance(label, list):
+                    if len(label) == n_csets:
+                        self._cslabels = list(label)
+
+                    else:
+                        self._cslabels = [''] * n_csets
+                        LOGGER.warn('Number of labels does not match number '
+                                    'of coordinate sets.')
+                else:
+                    self._cslabels = [str(label)] * n_csets
+            self._acsi = 0
+            self._setTimeStamp()
+
+        else:
+            acsi = self._acsi
+            if ndim == 2:
+                self._anisous[acsi] = anisous
+            else:
+                self._anisous[acsi] = anisous[0]
+            self._setTimeStamp(acsi)
+            self._cslabels[acsi] = str(label)
+
+    def addCoordset(self, coords, label=None, anisous=None):
         """Add a coordinate set.  *coords* argument may be an object with
         :meth:`getCoordsets` method."""
 
@@ -594,8 +698,13 @@ def addCoordset(self, coords, label=None):
         if coords.ndim == 2:
             coords = coords.reshape((1, n_atoms, 3))
 
+        if anisous is not None and anisous.ndim == 2:
+            anisous = anisous.reshape((1, n_atoms, 6))
+
         diff = coords.shape[0]
         self._coords = np.concatenate((self._coords, coords), axis=0)
+        if anisous is not None and self._anisous is not None:
+            self._anisous = np.concatenate((self._anisous, anisous/10000), axis=0)
         self._n_csets = self._coords.shape[0]
         timestamps = self._timestamps
         self._timestamps = np.zeros(self._n_csets)

prody/atomic/fields.py

@@ -123,8 +123,6 @@ def getDocstr(self, meth, plural=True, selex=True):
     'altloc':    Field('altloc', DTYPE + '1',
                        doc='alternate location indicator',
                        selstr=('altloc A B', 'altloc _'),),
-    'anisou':    Field('anisou', float, doc='anisotropic temperature factor',
-                       ndim=2),
     'chain':     Field('chain', DTYPE + '6',  doc='chain identifier',
                        meth='Chid', none=HVNONE, synonym='chid',
                        selstr=('chain A', 'chid A B C', 'chain _')),

prody/atomic/select.py

@@ -1341,6 +1341,7 @@ def _and2(self, sel, loc, tokens, subset=None):
         isDataLabel = atoms.isDataLabel
         append = None
         wasand = False
+        wasdata = False
         while tokens:
             # check whether token is an array to avoid array == str comparison
             token = tokens.pop(0)
@@ -1354,9 +1355,10 @@ def _and2(self, sel, loc, tokens, subset=None):
                             .format(repr('and ... and')), ['and', 'and'])
                     append = None
                     wasand = True
+                    wasdata = False
                     continue
 
-                elif isFlagLabel(token):
+                elif isFlagLabel(token) and not wasdata:
                     flags.append(token)
                     append = None
 
@@ -1396,10 +1398,12 @@ def _and2(self, sel, loc, tokens, subset=None):
                         evals.append([])
                         append = evals[-1].append
                     append(token)
+                    wasdata = True
 
                 elif token in UNARY:
                     unary.append([])
                     append = unary[-1].append
+                    wasdata = False
 
                     if token == 'not':
                         append((token,))
@@ -2420,3 +2424,10 @@ def _getCoords(self):
         if self._coords is None:
             self._coords = self._atoms._getCoords()
         return self._coords
+
+    def _getAnisous(self):
+        """Returns anisotropic temperature factors of atoms."""
+
+        if self._anisous is None:
+            self._anisous = self._atoms._getAnisous()
+        return self._anisous

prody/atomic/subset.py

@@ -63,13 +63,29 @@ def getCoords(self):
 
     _getCoords = getCoords
 
+    def getAnisous(self):
+        """Returns a copy of anisotropic temperature factors from the active coordinate set."""
+
+        if self._ag._anisous is not None:
+            # Since this is not slicing, a view is not returned
+            return self._ag._anisous[self.getACSIndex(), self._indices]
+
+    _getAnisous = getAnisous
+
     def setCoords(self, coords):
         """Set coordinates in the active coordinate set."""
 
         if self._ag._coords is not None:
             self._ag._coords[self.getACSIndex(), self._indices] = coords
             self._ag._setTimeStamp(self.getACSIndex())
 
+    def setAnisous(self, anisous):
+        """Set anisotropic temperature factors in the active coordinate set."""
+
+        if self._ag._anisous is not None:
+            self._ag._anisous[self.getACSIndex(), self._indices] = anisous
+            self._ag._setTimeStamp(self.getACSIndex())
+
     def getCoordsets(self, indices=None):
         """Returns coordinate set(s) at given *indices*, which may be an integer
         or a list/array of integers."""

prody/database/__init__.py

@@ -10,8 +10,7 @@
   * :func:`.fetchPfamMSA` - download MSA files
   * :func:`.searchPfam` - search for domain families of a protein
 
-
-.. _Pfam: http://pfam.sanger.ac.uk/
+.. _Pfam: https://www.ebi.ac.uk/interpro/entry/pfam/
 
 UniProt
 ========
@@ -70,7 +69,14 @@
 
 .. _GOA: https://www.ebi.ac.uk/GOA/
 
+Interpro
+====
+
+The following functions can be used to search and retrieve Pfam_ data:
+
+  * :func:`.searchInterpro` - search for domain families of a protein
 
+.. _Pfam: https://www.ebi.ac.uk/interpro/
 """
 
 __all__ = []
@@ -98,3 +104,7 @@
 from . import quartataweb
 from .quartataweb import *
 __all__.extend(quartataweb.__all__)
+
+from . import interpro
+from .interpro import *
+__all__.extend(interpro.__all__)