Closes #71 MSA is no longer based on TransposonLayout. Removed 'trans…

…poson' fluentdna.layout option. Code cleanup. Ran full test suite.

DDV/AnnotationAlignment.py

@@ -18,14 +18,9 @@
     filter_repeats_by_chromosome_and_family
 from DDV.Span import alignment_chopping_index, AlignedSpans, Span
 from DDV.TransposonLayout import TransposonLayout
-
 from DDV.DDVUtils import make_output_dir_with_suffix
 
 
-# class AnnotationAlignment(object):
-#     def __init__(self):
-
-
 def create_aligned_annotation_fragments(alignment, repeat_entries):
     aligned_repeats = []  # list of AlignedSpans
     consensus_width = max_consensus_width(repeat_entries)

DDV/MultipleAlignmentLayout.py

@@ -7,10 +7,9 @@
 from PIL import Image, ImageDraw
 
 import math
-from DDV.TileLayout import hex_to_rgb
+from DDV.TileLayout import hex_to_rgb, TileLayout
 from natsort import natsorted
 
-from DDV.TransposonLayout import TransposonLayout
 from Layouts import level_layout_factory
 
 
@@ -24,13 +23,14 @@ def fastas_in_folder(input_fasta_folder):
 
 
 
-class MultipleAlignmentLayout(TransposonLayout):
+class MultipleAlignmentLayout(TileLayout):
     def __init__(self, sort_contigs=False, **kwargs):
         kwargs['low_contrast'] = True
         kwargs['sort_contigs'] = True
         super(MultipleAlignmentLayout, self).__init__(**kwargs)
         self.all_contents = {}  # (filename: contigs) output_fasta() determines order of fasta_sources
-        self.using_mixed_widths = True  # we are processing all repeat types with different widths
+        self.current_column_height = 20
+        self.next_origin = [self.border_width, 30] # margin for titles, incremented each MSA
         self.protein_palette = True
         self.sort_contigs = sort_contigs
         self.title_height_px = 10
@@ -90,7 +90,6 @@ def __init__(self, sort_contigs=False, **kwargs):
 
 
     def process_all_alignments(self, input_fasta_folder, output_folder, output_file_name):
-        self.using_mixed_widths = True  # we are processing all repeat types with different widths
         start_time = datetime.now()
         self.preview_all_files(input_fasta_folder)
         self.calculate_mixed_layout()
@@ -115,11 +114,11 @@ def process_all_alignments(self, input_fasta_folder, output_folder, output_file_
         print("Output Image in:", datetime.now() - start_time)
 
 
-    def draw_nucleotides(self):
+    def draw_nucleotides(self, verbose=False):
         """Layout a whole set of different repeat types with different widths.  Column height is fixed,
         but column width varies constantly.  Wrapping to the next row is determined by hitting the
         edge of the allocated image."""
-        super(TransposonLayout, self).draw_nucleotides()
+        super(MultipleAlignmentLayout, self).draw_nucleotides(verbose)
 
 
     def calc_all_padding(self):
@@ -152,8 +151,9 @@ def prepare_image(self, image_length, width=None, height=None):
         self.pixels = self.image.load()
 
     def initialize_image_by_sequence_dimensions(self):
-        max_w = max([x.consensus_width for source in self.all_contents.values() for x in source])
-        max_h = max([len(source) for source in self.all_contents.values()])
+        margin = self.border_width*2
+        max_w = max([x.consensus_width for source in self.all_contents.values() for x in source]) + margin
+        max_h = max([len(source) + self.title_height_px for source in self.all_contents.values()]) + margin
         #TODO check if max_h is too large and needs to be interrupted by layout: one full row
         areas = []
         for source in self.all_contents.values():

DDV/TileLayout.py

@@ -222,7 +222,7 @@ def draw_extras(self):
         """Placeholder method for child classes"""
         pass
 
-    def draw_nucleotides(self):
+    def draw_nucleotides(self, verbose=True):
         total_progress = 0
         # Layout contigs one at a time
         for contig_index, contig in enumerate(self.contigs):
@@ -239,9 +239,9 @@ def draw_nucleotides(self):
                         # if nuc != gap_char:
                         self.draw_pixel(nuc, x + i, y)
                 except IndexError:
-                   print("Cursor fell off the image at", x,y)
+                   print("Cursor fell off the image at", (x,y))
             total_progress += contig.tail_padding  # add trailing white space after the contig sequence body
-            if len(self.contigs) < 100 or contig_index % (len(self.contigs) // 100) == 0:
+            if verbose and (len(self.contigs) < 100 or contig_index % (len(self.contigs) // 100) == 0):
                 print(str(total_progress / self.image_length * 100)[:4], '% done:', contig.name,
                       flush=True)  # pseudo progress bar
         print('')

DDV/TransposonLayout.py

@@ -1,9 +1,19 @@
+""" DEPRECATION WARNING:
+This file is currently unused by FluentDNA.  The original purpose was to show
+a variety of MSA pulled from RepeatMasker annotations across the genome.
+The intent was to visualize the diversity and abundance at each site.
+All the reusable functionality of TransposonLayout has been moved to MultipleAlignmentLayout.
+A fairly small script could process the repeatMasker annotation file into a folder of fasta MSA
+that could be visualized with MultipleAlignmentLayout.
+The crucial values are the within repeat coordinates rep_end.  I found experimentally
+that rooting the MSA on the last nucleotide of each line (not the start) gave the most
+coherent MSA."""
+
 from __future__ import print_function, division, absolute_import, \
     with_statement, generators, nested_scopes
 import math
 import traceback
 
-import sys
 from DNASkittleUtils.DDVUtils import editable_str
 from collections import defaultdict
 from datetime import datetime
@@ -12,15 +22,14 @@
 from DNASkittleUtils.DDVUtils import rev_comp
 from DDV.RepeatAnnotations import read_repeatmasker_csv, max_consensus_width, blank_line_array
 from DDV.TileLayout import TileLayout
-from DDV.Layouts import LayoutLevel, level_layout_factory
 from DDV import gap_char
 
 
 class TransposonLayout(TileLayout):
     def __init__(self, **kwargs):
         # print("Warning: Transposon Layout is an experimental feature not currently supported.",
         #       file=sys.stderr)
-        kwargs.update({'sort_contigs': True})  # important for mega row heigh handling
+        kwargs.update({'sort_contigs': True})  # important for mega row height handling
         super(TransposonLayout, self).__init__(**kwargs)
         self.repeat_entries = None
         self.current_column_height = 20
@@ -102,7 +111,7 @@ def filter_simple_repeats(self, return_only_simple_repeats=False):
 
 
 
-    def draw_nucleotides(self):
+    def draw_nucleotides(self, verbose=True):
         processed_contigs = self.create_repeat_fasta_contigs()
         print("Finished creating contigs")
         self.contigs = processed_contigs  # TODO: overwriting self.contigs isn't really great data management

DDV/fluentdna.py

@@ -55,7 +55,6 @@
 from DDV.UniqueOnlyChainParser import UniqueOnlyChainParser
 from DDV.AnnotatedAlignment import AnnotatedAlignment
 from DDV.TileLayout import TileLayout
-from DDV.TransposonLayout import TransposonLayout
 from DDV.MultipleAlignmentLayout import MultipleAlignmentLayout
 from DNASkittleUtils.Contigs import write_contigs_to_file, read_contigs
 
@@ -150,14 +149,6 @@ def ddv(args):
         done(args, args.output_dir)
 
     # ==========TODO: separate views that support batches of contigs============= #
-    elif args.layout == 'transposon':
-        layout = TransposonLayout()
-        # if len(args.contigs) != 1:
-        #     raise NotImplementedError("Chromosome Argument requires exactly one chromosome e.g. '--contigs chr12'")
-        layout.process_all_repeats(args.fasta, args.output_dir, just_the_name(args.output_dir), args.ref_annotation, args.contigs)
-        print("Done with Transposons")
-        done(args, args.output_dir)
-
     elif args.layout == 'alignment':
         layout = MultipleAlignmentLayout(sort_contigs=args.sort_contigs)
         layout.process_all_alignments(args.fasta,