Add the hydrophilicity scale and descriptors from Barley (2018)

README.md

 # `peptides.py` [![Stars](https://img.shields.io/github/stars/althonos/peptides.py.svg?style=social&maxAge=3600&label=Star)](https://github.com/althonos/peptides.py/stargazers)
 
-*Physicochemical properties and indices for amino-acid sequences.*
+*Physicochemical properties, indices and descriptors for amino-acid sequences.*
 
 [![Actions](https://img.shields.io/github/workflow/status/althonos/peptides.py/Test/main?logo=github&style=flat-square&maxAge=300)](https://github.com/althonos/peptides.py/actions)
 [![Coverage](https://img.shields.io/codecov/c/gh/althonos/peptides.py?style=flat-square&maxAge=3600)](https://codecov.io/gh/althonos/peptides.py/)
@@ -36,7 +36,7 @@ $ pip install peptides
 ```
 
 <!--
-Otherwise, Pyrodigal is also available as a [Bioconda](https://bioconda.github.io/)
+Otherwise, `peptides.py` is also available as a [Bioconda](https://bioconda.github.io/)
 package:
 ```console
 $ conda install -c bioconda peptides-py

docs/index.rst

@@ -9,7 +9,7 @@
 .. |Stars| image:: https://img.shields.io/github/stars/althonos/peptides.py.svg?style=social&maxAge=3600&label=Star
    :target: https://github.com/althonos/peptides.py/stargazers
 
-*Physicochemical properties and indices for amino-acid sequences.*
+*Physicochemical properties, indices and descriptors for amino-acid sequences.*
 
 |Actions| |Coverage| |PyPI| |Wheel| |Versions| |Implementations| |License| |Source| |Mirror| |Issues| |Docs| |Changelog| |Downloads|
 
@@ -88,7 +88,7 @@ A non-exhaustive list of available features:
 
 - Sequence profiles:
 
-  - Hydrophobicity profile using one of 38 proposed scales.
+  - Hydrophobicity profile using one of 39 proposed scales.
   - Hydrophobic moment profile based on `Eisenberg, Weiss and Terwilliger (1984) <https://doi.org/10.1073/pnas.81.1.140>`_.
   - Membrane position based on `Eisenberg (1984) <https://doi.org/10.1146/annurev.bi.53.070184.003115>`_.
 

peptides/__init__.py

@@ -69,6 +69,11 @@ class MSWHIMScores(typing.NamedTuple):
     mswhim3: float
 
 
+class PhysicalDescriptors(typing.NamedTuple):
+    pd1: float
+    pd2: float
+
+
 class PCPDescriptors(typing.NamedTuple):
     e1: float
     e2: float
@@ -617,7 +622,7 @@ class Peptide(typing.Sequence[str]):
 
         This function calculates the hydrophobicity index of an amino
         acid sequence by averaging the hydrophobicity values of each residue
-        using one of the 38 scales from different sources.
+        using one of the 39 scales from different sources.
 
         Arguments:
             scale (`str`): The name of the hydrophobicity scale to be used.
@@ -654,10 +659,15 @@ class Peptide(typing.Sequence[str]):
               *Structural Prediction of Membrane-Bound Proteins*.
               European Journal of Biochemistry. Nov 1982;128(2–3):565–75.
               doi:10.1111/j.1432-1033.1982.tb07002.x. PMID:7151796.
+            - Barley, M. H., N. J. Turner, and R. Goodacre.
+              *Improved Descriptors for the Quantitative Structure–Activity
+              Relationship Modeling of Peptides and Proteins*. Journal of
+              Chemical Information and Modeling. Feb 2018;58(2):234–43.
+              doi:10.1021/acs.jcim.7b00488. PMID:29338232.
             - Black, S. D., and D. R. Mould.
               *Development of Hydrophobicity Parameters to Analyze Proteins
               Which Bear Post- or Cotranslational Modifications*.
-              Analytical Biochemistry. Feb 1991;193(1):72-82.
+              Analytical Biochemistry. Feb 1991;193(1):72–82.
               doi:10.1016/0003-2697(91)90045-u. PMID:2042744.
             - Bull, H. B., and K. Breese.
               *Surface Tension of Amino Acid Solutions: A Hydrophobicity
@@ -1461,7 +1471,7 @@ class Peptide(typing.Sequence[str]):
 
         Returns:
             `peptides.PCPDescriptors`: The computed average of PCP
-            descriptors descriptors of all the amino acids in the peptide.
+            descriptors of all the amino acids in the peptide.
 
         Example:
             >>> peptide = Peptide("QWGRRCCGWGPGRRYCVRWC")
@@ -1495,6 +1505,53 @@ class Peptide(typing.Sequence[str]):
             )
         return PCPDescriptors(*out)
 
+    def physical_descriptors(self) -> PhysicalDescriptors:
+        """Compute the Physical Descriptors of a protein sequence.
+
+        The PP descriptors were constructed by improving on existing
+        PCA-derived descriptors (Z-scales, MS-WHIM and T-scales) after
+        correcting for the hydrophilicity of Methionine, Asparagine and
+        Tryptophan based on Feng *et al*.
+
+        Returns:
+            `peptides.PhyiscalDescriptors`: The computed average of Physical
+            Descriptors of all the amino acids in the peptide. *PD1* is
+            related to volume while *PD2* is related to hydrophilicity.
+
+        Example:
+            >>> peptide = Peptide("QWGRRCCGWGPGRRYCVRWC")
+            >>> for i, pd in enumerate(peptide.physical_descriptors()):
+            ...     print(f"PD{i+1:<3} {pd: .4f}")
+            PD1    0.1190
+            PD2    0.2825
+
+        Note:
+            Barley *et al* insisted on maintaining a minimal number of
+            descriptors as a way to reduce the chances of finding spurious
+            QSAM models that would be affected by mutation between
+            interaction sites.
+
+        References:
+          - Barley, M. H., N. J. Turner, and R. Goodacre.
+            *Improved Descriptors for the Quantitative Structure–Activity
+            Relationship Modeling of Peptides and Proteins*. Journal of
+            Chemical Information and Modeling. Feb 2018;58(2):234–43.
+            doi:10.1021/acs.jcim.7b00488. PMID:29338232.
+          - Feng, X., J. Sanchis, M. T. Reetz, and H. Rabitz.
+            *Enhancing the Efficiency of Directed Evolution in Focused
+            Enzyme Libraries by the Adaptive Substituent Reordering
+            Algorithm*. Chemistry. Apr 2012;18(18):5646–54.
+            doi:10.1002/chem.201103811. PMID:22434591.
+
+        """
+        out = array.array("d")
+        for i in range(len(tables.PHYSICAL_DESCRIPTORS)):
+            scale = tables.PHYSICAL_DESCRIPTORS[f"PD{i+1}"]
+            out.append(
+                sum(scale.get(aa, 0) for aa in self.sequence) / len(self.sequence)
+            )
+        return PhysicalDescriptors(*out)
+
     def protfp_descriptors(self) -> ProtFPDescriptors:
         """Compute the ProtFP descriptors of a protein sequence.
 

peptides/data/hydrophobicity/Barley.csv

+../physical_descriptors/PD2.csv
\ No newline at end of file

peptides/data/physical_descriptors/PD1.csv

+A,-2.90
+R,2.41
+N,-0.68
+D,-0.92
+C,-1.89
+Q,0.36
+E,0.16
+G,-4.04
+H,0.83
+I,0.51
+L,0.52
+K,0.92
+M,0.92
+F,2.22
+P,-1.25
+S,-2.36
+T,-1.19
+W,4.28
+Y,2.75
+V,-0.65

peptides/data/physical_descriptors/PD2.csv

+A,-1.03
+R,1.31
+N,0.79
+D,1.23
+C,0.15
+Q,1.09
+E,1.28
+G,0.01
+H,1.15
+I,-1.32
+L,-1.40
+K,1.23
+M,-1.42
+F,-1.47
+P,-0.64
+S,0.38
+T,0.28
+W,-0.18
+Y,-0.18
+V,-1.27

setup.cfg

@@ -8,7 +8,7 @@ version = attr: peptides.__version__
 author = Martin Larralde
 author_email = martin.larralde@embl.de
 url = https://github.com/althonos/peptides.py
-description = Physicochemical properties and indices for amino-acid sequences (ported from R).
+description = Physicochemical properties, indices and descriptors for amino-acid sequences.
 long_description = file: README.md
 long_description_content_type = text/markdown
 license = MIT