support info_annotations for BED variant files

README.md

@@ -13,7 +13,7 @@ Tracks are specified as GTF or BED and are annotated with name, coordinates, and
 
 Coordinate-based information can be provided as CSV/TSV (point-based) or BEDGRAPH (interval-based). The user can customize the y axis with tick precision and scientific notation. The user can specify the line color, alpha value, and choose whether to fill in the area under the curve.
 
-Headers are not supported for BED files. BED files must have at least three columns, which should be `chrom`, `start`, and `end`.
+If BED file contains a header, the line must begin with `#`. BED files must have at least three columns, which should be `chrom`, `start`, and `end`.
 
 Plots can be output as HTML, PNG, or SVG.
 
@@ -76,7 +76,7 @@ https://home.chpc.utah.edu/~u6038618/transcriptionary/plot.html
 - `format`: file format of variant file; 'vcf'
 - `filepath`: path to VCF
 - `chrom`: chromosome
-- `info_annotations`: VCF only; INFO fields to add to hover boxes
+- `info_annotations`: for VCF, INFO fields to add to hover boxes
     - `<info_field_1>`
 - `vep`: VCF only; leave empty if not VEP annotated
     - `field_name`: name of INFO field with VEP string (e.g. vep, ann, csq)
@@ -89,12 +89,14 @@ https://home.chpc.utah.edu/~u6038618/transcriptionary/plot.html
     - `MED`:
     - `HIGH`:
 
-`#BED (no header)`
+`#BED (if header in file, it must start with #)`
 `<variant_set>`: variant set label
 - `format`: file format of variant file; 'vcf'
 - `filepath`: path to VCF
 - `chrom`: chromosome
 - `header`: BED only; list of column names in BED file
+- `info_annotations`: for BED, column names to add to hover boxes
+        - `<info_field_1>`
 - `color`: default lollipop color; use hex codes or predefined colors from default_colors/named_colors.yaml
 - `variant_severity_colors`: specify lollipop colors by variant severity; use hex codes or predefined colors from default_colors/named_colors.yaml
     - `LOW`:

transcriptionary/get_coords.py

@@ -88,14 +88,17 @@ def get_variants_bed(bed_path):
             if fi.readline().count('\t') != len(variant_params[variant_set]['header']) - 1:
                 print('Warning: number of fields in variant set {} header argument not equal to number of fields in file {} (check for trailing delimiters).'.format(variant_set, bed_path))
 
-        df = pd.read_csv(bed_path, names=variant_params[variant_set]['header'], sep='\t', index_col=None) 
+        df = pd.read_csv(bed_path, names=variant_params[variant_set]['header'], sep='\t', index_col=None, keep_default_na=False, comment='#') 
         variant_params[variant_set]['has_yaxis_info'] = False
+        if variant_params[variant_set]['consequence_idx']: plot_params['add_variant_severity_checkbox'] = True
 
         for _,row in df.iterrows():
             if str(row.iloc[0]) != str(variant_params[variant_set]['chrom']): continue
             di_variant = dict(pos=row.iloc[1], compact_start=-1, variant_set=variant_set, info_annotations=variant_params[variant_set]['info_annotations'], vep=variant_params[variant_set]['vep'], allele_count=0, allele_frequency=0, allele_number=0)
+
+            for info_field in variant_params[variant_set]['info_annotations']: di_variant[info_field] = row[info_field]
+
             if variant_params[variant_set]['consequence_idx']: #0-based
-                plot_params['add_variant_severity_checkbox'] = True
                 for transcript_ID in transcript_IDs: 
                     di_variant[transcript_ID + '_severity'] = VEP(row.iloc[variant_params[variant_set]['consequence_idx']],['Consequence']).impact_severity
             else:

transcriptionary/transcriptionary.py

@@ -2,7 +2,7 @@
 from .get_coords import get_variants,get_line,get_track
 from .colors import color_boxes
 from .axes import add_user_axis,add_variant_axis
-from .glyphs import add_intron_glyph, add_exon_glyph, add_variant_glyph, add_UTR_glyph, add_track_glyph, add_multi_line_glyph, flatten
+from .glyphs import add_intron_glyph, add_exon_glyph, add_variant_glyph, add_UTR_glyph, add_track_glyph, add_multi_line_glyph
 from .widget_callbacks import add_checkbox,add_variant_severity_checkbox,add_user_tracks_checkbox,add_user_lines_checkbox,add_smoothing_slider,add_legend,add_linear_log_scale,add_exon_zoom,add_variant_sets_checkbox
 from . import project_coords
 import numpy as np
@@ -55,7 +55,7 @@ def plot_transcript(plot_params, variant_params, user_track_params, user_line_pa
     if plot_params['add_variant_axis']:
         def log10(f): return np.log10(f) if f > 0 else 0
 
-        all_vars = flatten([variant_params[var_set]['variant_ls'] for var_set in variant_params]) 
+        all_vars = sum([variant_params[var_set]['variant_ls'] for var_set in variant_params], []) #flatten list 
         all_vars_in_transcript = [v for v in all_vars if v['compact_pos'] >= 0] #get list of all variants in transcript to get max and min for variant axes
         allele_counts = [v['allele_count'] for v in all_vars_in_transcript]
         allele_frequencies = [v['allele_frequency'] for v in all_vars_in_transcript]
@@ -114,7 +114,7 @@ def log10(f): return np.log10(f) if f > 0 else 0
                 xs_ls,ys_ls = project_coords.map_line(user_lines[axis_name][line], transcript_dict['exons'], user_line_params[axis_name]['lines'][line]['chrom'])
                 all_xs.append(xs_ls)
                 all_ys.append(ys_ls)
-        try: y_max = max([i for s in [i for s in all_ys for i in s] for i in s]) #flatten 3D list to 1D list to take max #TODO change to flatten
+        try: y_max = max([i for s in [i for s in all_ys for i in s] for i in s]) #flatten 3D list to 1D list to take max
         except: continue
 
         for idx,line in enumerate(user_line_params[axis_name]['lines']):
@@ -188,10 +188,19 @@ def get_color(color): return color if color[0] == "#" else named_colors[color]
                 variant_params[variant_set]['info_annotations'] = []
         except: variant_params[variant_set]['info_annotations'] = []
 
+        try: #if consequence_idx nonexistent,
+            int(variant_params[variant_set]['consequence_idx'])
+        except: variant_params[variant_set]['consequence_idx'] = False
+
+        #for BED files, replace space with underscore in annotation field names (space causes issues with hover box)
+        try:
+            variant_params[variant_set]['header'] = list(map(lambda field_name: str.replace(field_name, ' ', '_'), variant_params[variant_set]['header']))
+            variant_params[variant_set]['info_annotations'] = list(map(lambda field_name: str.replace(field_name, ' ', '_'), variant_params[variant_set]['info_annotations']))
+        except: pass
+
         try: #if vep empty or nonexistent, set to empty params
             if not variant_params[variant_set]['vep']: variant_params[variant_set]['vep'] = {'field_name': '' ,'vep_fields': [], 'annotate_severity_by': ''}
         except: variant_params[variant_set]['vep'] = {'field_name': '' ,'vep_fields': [], 'annotate_severity_by': ''}
-        # except: variant_params[variant_set]['vep'] = {'vep_fields': []}
 
     ### TRACKS ###
     for track_name in user_track_params:
@@ -249,7 +258,6 @@ def transcriptionary():
     for line_name in user_line_params:
         line_axes[line_name] = []
 
-    #for idx,ID in enumerate(transcript_IDs):
     for ID in transcript_IDs:
         title = 'gene={}; transcript={}/{}'.format(plot_params['gene_name'], ID, transcripts[ID]['ID'])
         if transcripts[ID]['direction']: title += ' ({})'.format(transcripts[ID]['direction'])