Add bioplib speed comparison

Co-authored-by: AndrewCRMartin <andrew@bioinf.org.uk>

.gitignore

@@ -11,3 +11,5 @@ profiles
 *.png
 *.json
 *.svg
+*.eps
+readapdb

scripts/readapdb.c

@@ -0,0 +1,32 @@
+/*
+
+This code written by Professor Andrew Martin on 2019-08-08
+
+alias cc='/usr/bin/cc -I$HOME/include -L$HOME/lib -ansi -pedantic -Wall'
+cc -o readapdb readapdb.c -lbiop -lgen -lm -lxml2
+*/
+
+#include <stdio.h>
+#include "bioplib/pdb.h"
+
+int main(int argc, char **argv)
+{
+   FILE *fp;
+
+   if((fp=fopen(argv[1], "r"))!=NULL)
+   {
+      int natoms;
+      PDB *pdb;
+
+      if((pdb = blReadPDB(fp, &natoms))!=NULL)
+      {
+
+         PDBSTRUCT *pdbs;
+         pdbs = blAllocPDBStructure(pdb);
+
+      }
+   }
+
+   return(0);
+}
+

scripts/speed.py

@@ -1,5 +1,6 @@
 import requests
 import random
+import subprocess
 import sys
 import os
 import json
@@ -32,7 +33,7 @@ def get_string(code):
     codes = response.text.split()
 
     data = []
-    while len (data) != 1000:
+    while len (data) != 10:
         code = random.choice(codes)
         d = {"code": code}
         print(len(data) + 1, code)
@@ -61,13 +62,26 @@ def get_string(code):
                 d["biopython"] = delta.total_seconds()
             except:
                 d["biopython"] = None
+
+            try:
+                with open("temp.pdb", "w") as f: f.write(string)
+                start = datetime.now()
+                subprocess.check_output("scripts/readapdb temp.pdb", shell=True)
+                end = datetime.now()
+                delta = end - start
+                d["bioplib"] = delta.total_seconds()
+            except:
+                d["bioplib"] = None
 
         else:
             d["biopython"] = None
+            d["bioplib"] = None
 
         data.append(d)
         codes.remove(code)
 
+        if os.path.exists("temp.pdb"):
+            os.remove("temp.pdb")
         with open("scripts/speed.json", "w") as f:
             json.dump(data, f)
 
@@ -81,21 +95,17 @@ def get_string(code):
 cifs_x, cifs_y = zip(*[[d["atoms"], d[".cif"]] for d in data if d[".cif"] and d["models"] == 1 and d[".cif"] < 20])
 mmtfs_x, mmtfs_y = zip(*[[d["atoms"], d[".mmtf"]] for d in data if d[".mmtf"] and d["models"] == 1 and d[".mmtf"] < 20])
 bios_x, bios_y = zip(*[[d["atoms"], d["biopython"]] for d in data if d["biopython"] and d["models"] == 1 and d["biopython"] < 20])
+bioplib_x, bioplib_y = zip(*[[d["atoms"], d["bioplib"]] for d in data if d["bioplib"] and d["models"] == 1 and d["bioplib"] < 20])
 
 def best_fit(X, Y, label):
-
     xbar = sum(X)/len(X)
     ybar = sum(Y)/len(Y)
     n = len(X) # or len(Y)
-
     numer = sum([xi*yi for xi,yi in zip(X, Y)]) - n * xbar * ybar
     denum = sum([xi**2 for xi in X]) - n * xbar**2
-
     b = numer / denum
     a = ybar - b * xbar
-
-    print('{} best fit line:\ny = {:.6f} + {:.6f}x'.format(label, a, b))
-
+    print("{} best fit line:\ny = {:.6f} + {:.6f}x".format(label, a, b))
     return a, b
 
 plt.xscale("log")
@@ -104,6 +114,7 @@ def best_fit(X, Y, label):
 best_fit(pdbs_x, pdbs_y, "pdb")
 best_fit(mmtfs_x, mmtfs_y, "mmtf")
 best_fit(bios_x, bios_y, "biopython")
+best_fit(bioplib_x, bioplib_y, "bioplib")
 plt.scatter(cifs_x, cifs_y, s=6, c="#FD7272", label=".cif", alpha=0.3, linewidths=0)
 plt.scatter(pdbs_x, pdbs_y, s=6, c="#58B19F", label=".pdb", alpha=0.3, linewidths=0)
 plt.scatter(mmtfs_x, mmtfs_y, s=6, c="#182C61", label=".mmtf", alpha=0.3, linewidths=0)
@@ -112,22 +123,20 @@ def best_fit(X, Y, label):
 plt.xlim([100, 1000000])
 plt.ylim([0.001, 100])
 plt.legend(loc=2)
-plt.savefig("scripts/format-speed.svg", dpi=1000)
+plt.savefig("scripts/format-speed.png", dpi=1000)
 plt.clf()
 
 
 plt.xscale("log")
 plt.yscale("log")
 plt.scatter(pdbs_x, pdbs_y, s=6, c="#58B19F", label=".pdb (atomium)", alpha=0.5, linewidths=0)
 plt.scatter(bios_x, bios_y, s=6, c="#D6A2E8", label=".pdb (biopython)", alpha=0.5, linewidths=0)
+plt.scatter(bioplib_x, bioplib_y, s=6, c="#333333", label=".pdb (bioplib)", alpha=0.5, linewidths=0)
 plt.xlabel("Atom Count")
 plt.ylabel("Parse time (s)")
 plt.xlim([100, 100000])
 plt.ylim([0.001, 10])
 plt.legend(loc=2)
-plt.savefig("scripts/library-speed.svg", dpi=1000)
+plt.savefig("scripts/library-speed.png", dpi=1000)
 plt.clf()
 
-
-#plt.scatter(bios_x, bios_y, s=4, c="#F97F51")
-