Update v1.39

README.md

@@ -75,6 +75,14 @@ Write all values of a column to a file. For example, passing *\_rlnMicrographNam
 
 Find particles that are shared between the input star file and the one provided by ```--f``` based on the column provided here. Two new star files will be output, one with the shared particles and one with the unique particles.
 
+**```--match_mics```**
+
+FKeep only micrographs that also exist in a second star file provided by ```--f```.
+
+**```--extract_min```** *```minimum-value```*
+
+Find the micrographs that have this minimum number of particles in them and extract all the particles belonging to them.
+
 **```--extract_if_nearby```** *```distance```* *`--f otherfile.star`*
 
 For every particle in the input star file, check the nearest particle in a second star file provided by ```--f```; particles that have a neighbor closer than the distance (in pixels) provided here will be written to particles_close.star, and those that don't will be written to particles_far.star. Particles that couldn't be matched to a neighbor will be skipped (i.e. if the second star file lacks particles in that micrograph). It will also output a histogram of nearest distances to Particles_distances.png (use ```--t``` to change the file type; see the [*Output*](#output) options).

starparser/__init__.py

@@ -1,4 +1,4 @@
 import os
 
-__version__ = '1.38'
+__version__ = '1.39'
 _ROOT = os.path.abspath(os.path.dirname(__file__))

starparser/argparser.py

@@ -47,6 +47,14 @@ def argparse():
         action="store", dest="parser_findshared", type="string", default="", metavar='column-name',
         help="Find particles that are shared between the input star file and the one provided by --f based on the column provided here. Two new star files will be written, one with the shared particles and one with the unique particles.")
 
+    info_opts.add_option("--match_mics",
+        action="store_true", dest="parser_matchmics", default=False,
+        help="Keep only micrographs that also exist in a second star file provided by --f.")
+
+    info_opts.add_option("--extract_min",
+        action="store", dest="parser_exractmin", type="int", default=-1, metavar='minimum-number',
+        help="Find the micrographs that have this minimum number of particles in them and extract all the particles belonging to them.")
+
     info_opts.add_option("--extract_if_nearby",
         action="store", dest="parser_findnearby", type="float", default=-1, metavar='distance',
         help="Find the nearest particle in a second star file (specified by --f); particles that have a neighbor in the second star file closer than the distance provided here will be written to particles_close.star and those that don't will be written to particles_far.star. Particles that couldn't be matched to a neighbor will be skipped (i.e. if the second star file lacks particles in that micrograph). It will also write a histogram of nearest distances to Particles_distances.png.")

starparser/decisiontree.py

@@ -479,6 +479,26 @@ def decide():
             fileparser.writestar(unsharedparticles, metadata, "unique.star", relegateflag)
         sys.exit()
 
+
+    """
+    --match_mics
+    """
+
+    if params["parser_matchmics"]:
+        columntocheckunique = params["parser_findshared"]
+        if params["parser_file2"] == "":
+            print("\n>> Error: provide a second file with --f to compare to.\n")
+            sys.exit()
+        file2 = params["parser_file2"]
+        if not os.path.isfile(file2):
+            print("\n>> Error: \"" + file2 + "\" does not exist.\n")
+            sys.exit();
+        otherparticles, f2metadata = fileparser.getparticles(file2)
+        matchedparticles = allparticles[allparticles["_rlnMicrographName"].isin(otherparticles["_rlnMicrographName"])]
+        print("\n>> Kept " + str(len(set(matchedparticles["_rlnMicrographName"].tolist()))) + " micrographs that matched the second file (out of " + str(len(set(allparticles["_rlnMicrographName"].tolist()))) + ").\n")
+        fileparser.writestar(matchedparticles, metadata, "output.star", relegateflag)
+        sys.exit()
+
     """
     --extract_if_nearby
     """
@@ -567,6 +587,19 @@ def decide():
         fileparser.writestar(clusterparticles, metadata, params["parser_outname"], relegateflag)
         sys.exit()
 
+
+    """
+    --extract_min
+    """
+
+    if params["parser_exractmin"] != -1:
+        extractmin = params["parser_exractmin"]
+        print("\n>> Extracting particles that belong to micrographs with at least " + str(extractmin) + " particles.\n")
+        particlesfrommin = specialparticles.extractwithmin(allparticles, extractmin)
+        print(">> Removed " + str(len(allparticles.index)-len(particlesfrommin.index)) + " that did not match the criteria (" + str(len(particlesfrommin.index)) + " remaining out of " + str(len(allparticles.index)) + ").")
+        fileparser.writestar(particlesfrommin, metadata, params["parser_outname"], relegateflag)
+        sys.exit()
+
     """
     --plot_class_proportions
     """

starparser/specialparticles.py

@@ -270,4 +270,45 @@ def getcluster(particles,threshold,minimum):
 
     particles_purged = pd.concat(toconcat)
 
+    return(particles_purged)
+
+"""
+--extract_minimum
+"""
+def extractwithmin(particles,minimum):
+
+    #~needs explanation~#
+
+    uniquemics = particles.groupby(["_rlnMicrographName"])
+    nameloc = particles.columns.get_loc("_rlnImageName")+1
+
+    keep = []
+    badmics = 0
+    for mic in uniquemics:
+        if len(mic[1]) > minimum:
+            for particle in mic[1].itertuples():
+                keep.append(particle[nameloc])
+        else:
+            badmics+=1
+
+    if len(keep) == 0:
+        print("\n>> Error: no particles were retained based on the criteria.\n")
+        sys.exit()
+    elif len(keep) == len(particles.index):
+        print("\n>> Error: all particles were retained. No star file will be output.")
+        sys.exit()
+
+    print(">> " + str(badmics) + " micrographs don't meet the criteria.\n")
+
+    """
+    With dataframes, stating dataframe1 = dataframe2 only creates
+    a reference. Therefore, we must create a copy if we want to leave
+    the original dataframe unmodified.
+    """
+    particles_purged = particles.copy()
+
+    toconcat = [particles_purged[particles_purged["_rlnImageName"] == q] for q in keep]
+
+    particles_purged = pd.concat(toconcat)
+
     return(particles_purged)