annotate linear_genome_plot.py @ 4:318248986ee7 draft

planemo upload commit 4c7a6921e41a04d27f3ac202732bb5fb86801904-dirty

#!/usr/bin/env python
from Bio import SeqIO
from dna_features_viewer import BiopythonTranslator, GraphicRecord
from matplotlib import rc_context
import matplotlib
import matplotlib.pyplot as plt
from itertools import cycle
import re
import sys
import argparse


class CPTTranslator(BiopythonTranslator):
    """
    This is a customized translator from the dna_features_viewer module to fit Galaxy
    """

    global custom_feature_colors
    global box_status
    global label_fields
    global custom_name_colors
    global ignored_features_types
    global ignored_gene_labels
    global ignored_feature_labels

    def compute_feature_color(self, feature):
        if feature.type == "CDS":
            if "product" in feature.qualifiers:
                color_specific = any(
                    re.search(
                        ("(\\b" + str(item) + "\\b)"), feature.qualifiers["product"][0]
                    )
                    for item in custom_name_colors.keys()
                ) or any(
                    re.search((item), feature.qualifiers["product"][0])
                    for item in custom_name_colors.keys()
                )
                if color_specific:
                    try:
                        return custom_name_colors[feature.qualifiers["product"][0]]
                    except KeyError:
                        for item in custom_name_colors.keys():
                            if item in feature.qualifiers["product"][0]:
                                custom_name_colors[
                                    feature.qualifiers["product"][0]
                                ] = custom_name_colors[item]
                                return custom_name_colors[
                                    feature.qualifiers["product"][0]
                                ]
                                # print(feature.qualifiers["product"][0])
                else:
                    try:
                        return custom_feature_colors[feature.type]
                    except KeyError:
                        return BiopythonTranslator.compute_feature_color(self, feature)
        else:
            if feature.type not in ignored_features_types:
                try:
                    return custom_feature_colors[feature.type]
                except KeyError:
                    return BiopythonTranslator.compute_feature_color(self, feature)

    def compute_feature_label(self, feature):  # remove the chop_blocks
        self.label_fields = label_fields
        if feature.type == "CDS":
            if "product" in feature.qualifiers:
                if ignored_gene_labels:  #  product name drop
                    verify_chops = any(
                        re.search(
                            ("(\\b" + str(item) + "\\b)"),
                            feature.qualifiers["product"][0],
                        )
                        for item in ignored_gene_labels
                    ) or any(
                        re.search((item), feature.qualifiers["product"][0])
                        for item in ignored_gene_labels
                    )
                    if verify_chops:
                        return None
                    else:
                        return BiopythonTranslator.compute_feature_label(self, feature)
                else:
                    return BiopythonTranslator.compute_feature_label(self, feature)
        elif feature.type in ignored_feature_labels:
            return None
        else:
            return BiopythonTranslator.compute_feature_label(self, feature)

    def compute_filtered_features(self, features):
        return [
            feature
            for feature in features
            if feature.type not in ignored_features_types
        ]

    def compute_feature_legend_text(self, feature):
        return feature.type

    def compute_feature_box_color(self, feature):
        if feature.type == "CDS":
            return "white"

    def compute_feature_label_link_color(self, feature):
        return "black"

    def compute_feature_box_linewidth(self, feature):
        if box_status:
            return 0.5
        else:
            return 0


def parse_gbk(file):
    """simple function to parse out the feature information AND products"""

    record = SeqIO.read(file, "genbank")
    count = 0
    feature_types = {}
    product_names = []
    for feat in record.features:
        if feat.type not in feature_types:
            feature_types[feat.type] = 1
        else:
            feature_types[feat.type] += 1
        if "product" in feat.qualifiers:
            product_names.append(feat.qualifiers["product"][0])

    return feature_types, product_names, record


if __name__ == "__main__":
    parser = argparse.ArgumentParser(description="Linear Genome Plot")
    #  Input and Parameters
    parser.add_argument(
        "input_file", type=argparse.FileType("r"), help="genbank or gff3 file"
    )
    parser.add_argument("--plot_width", type=int, default=20)
    # parser.add_argument("--plot_height",type=int,default=4)
    parser.add_argument(
        "--title", type=str, default="genome plot"
    )  # NEED TO ADD TO XML
    parser.add_argument(
        "--common_features_excluded", default="", help="common features to be excluded"
    )
    parser.add_argument(
        "--features_excluded",
        default="",
        help="features to be excluded from plot, separate by commas",
    )
    parser.add_argument(
        "--common_ignore_feature_labels",
        default="",
        help="common feature labels to be excluded",
    )
    parser.add_argument(
        "--ignored_feature_labels",
        default="",
        help="ignore labeling of specific features",
    )
    parser.add_argument(
        "--common_ignore_product_labels",
        default="",
        help="common product names to not label",
    )
    parser.add_argument(
        "--ignore_labeling",
        default="",
        help="labeling for specific products to ignore, separate by commas",
    )
    parser.add_argument(
        "--feature_label_order",
        default="locus_tag",
        help="label order, where the first choice is the first feature listed to pull name labels from",
    )  # NEED TO ADD TO XML
    parser.add_argument(
        "--label_box",
        action="store_true",
        help="Use to have label box around feature labels",
    )
    parser.add_argument(
        "--label_algo",
        action="store_true",
        help="use dna features spacing algo for label placement (in or above feature)",
    )
    # parser.add_argument("--level_offset",type=int,default=0,help="All features and annotations will be pushed up by the input amount. Useful for when plotting several sets of features successively on the same axis.") # Will exclude for now
    # parser.add_argument("--custom_region",action="store_true",help="cropped region for plot")
    parser.add_argument("--sz", type=int, help="beginning location for crop")
    parser.add_argument("--ez", type=int, help="end location for crop")
    parser.add_argument("--st", type=int, help="start site of translation")
    parser.add_argument("--et", type=int, help="end site of translation")
    parser.add_argument(
        "--translation_on", action="store_true", help="plot the translation sub-axis"
    )
    parser.add_argument(
        "--feature_id",
        nargs="*",
        action="append",
        help="feature label to have custom color",
    )  # NEED TO ADD TO XML
    parser.add_argument(
        "--feature_id_color",
        nargs="*",
        action="append",
        help="feature's accompanying color",
    )
    parser.add_argument(
        "--gene_id",
        nargs="*",
        action="append",
        help="gene/cds label to have custom color",
    )
    parser.add_argument(
        "--gene_id_color",
        nargs="*",
        action="append",
        help="gene/cds's accompanying color",
    )
    parser.add_argument("--multiline", action="store_true", help="Plot multiline plot")
    parser.add_argument(
        "--nucl_per_line", type=int, help="nucleotides per line of multiline"
    )
    #  Output
    parser.add_argument(
        "--file_stats",
        type=argparse.FileType("w"),
        default="out_stats.txt",
        help="output stat file",
    )
    # parser.add_argument("--tmp_img",dest="tmp_img",type=argparse.FileType("wb"),default="out_tmp.svg")
    parser.add_argument(
        "--out_img",
        dest="out_img",
        type=argparse.FileType("wb"),
        default="out_img.svg",
        help="svg genome plot",
    )
    args = parser.parse_args()

    ##  Part I ; Parse and send output of features count and the list of product names
    feature_counts, products, genome = parse_gbk(args.input_file)
    with args.file_stats as f:
        f.writelines("---::: FILE BREAKDOWN :::---\n\n")
        f.writelines("------::: Feature Count :::------\n")
        for feature, count in feature_counts.items():
            f.writelines(f"Feature: {feature} ::::: Count: {count}\n")
        f.writelines("------::: Product Names :::------\n")
        if products != []:
            for product in products:
                f.writelines(f"Product Name: {product}\n")
        else:
            f.writelines("No Annotated Product Names Found")

    ##  Part II ; Prep Global Variables
    ##  Make K:V pairs for Feature Colors
    if args.label_box:
        box_status = True
    else:
        box_status = False

    if args.feature_id:
        feature_ids = [f for listed_obj in args.feature_id for f in listed_obj]
        feature_ids_colors = [
            f for listed_obj in args.feature_id_color for f in listed_obj
        ]
        custom_feature_colors = dict(zip(feature_ids, feature_ids_colors))
    else:
        custom_feature_colors = {}

    ##  Make K:V pairs for Name Colors (as above)
    if args.gene_id:
        gene_ids = [g for listed_obj in args.gene_id for g in listed_obj]
        gene_ids_colors = [g for listed_obj in args.gene_id_color for g in listed_obj]
        custom_name_colors = dict(zip(gene_ids, gene_ids_colors))
    else:
        custom_name_colors = {}

    ##  Ignored Features
    # ignored_features_types = str.split(args.features_excluded,",")
    if args.common_features_excluded:
        ignored_features_types = str.split(args.common_features_excluded, ",")
        if args.features_excluded:
            ignored_features_types += str.split(args.features_excluded, ",")
    elif args.features_excluded:
        ignored_features_types = str.split(args.features_excluded, ",")
    else:
        ignored_features_types = False

    print(ignored_features_types)

    ## product labels
    if args.common_ignore_product_labels:
        ignored_gene_labels = str.split(args.common_ignore_product_labels, ",")
        if args.ignore_labeling:
            ignored_gene_labels += str.split(args.ignore_labeling, ",")
    elif args.ignore_labeling:
        ignored_gene_labels = str.split(args.ignore_labeling, ",")
    else:
        ignored_gene_labels = False

    print(ignored_gene_labels)

    if args.feature_label_order != [""]:
        label_fields = str.split(args.feature_label_order, ",")

    # if ignored_gene_labels == ['']:
    #    ignored_gene_labels = False

    ##  Ignored Labeling
    if args.common_ignore_feature_labels:
        ignored_feature_labels = str.split(args.common_ignore_feature_labels, ",")
        if args.ignored_feature_labels:
            ignored_feature_labels += str.split(args.ignored_feature_labels, ",")
    elif args.ignored_feature_labels:
        ignored_feature_labels = str.split(args.ignored_feature_labels, ",")
    else:
        ignored_feature_labels = False

    print(ignored_feature_labels)
    ##  Print Statements for Debugging
    # print(custom_feature_colors)
    # print(custom_name_colors)
    # print(ignored_features_types)
    # print(ignored_gene_labels)
    # print(label_fields)

    ## Part III ; PLOT
    # Housekeeping
    rc_context(
        {
            "font.family": ["monospace"],
        }
    )  # courier-like
    matplotlib.use("Agg")  # I think this has to be used...

    if args.label_algo:
        lab_algo = True
    else:
        lab_algo = False

    translator = CPTTranslator()
    graphic_record = translator.translate_record(genome)

    with open("tmp.svg", "wb") as img:
        img.truncate(0)
        img.close()

    if (
        args.sz and not args.multiline
    ):  #  if user is wanting to look at a subset region of the genome
        zoom_start, zoom_end = args.sz, args.ez
        cropped = graphic_record.crop((zoom_start, zoom_end))
        ax, _ = cropped.plot(
            figure_width=args.plot_width, annotate_inline=lab_algo, figure_height=None
        )
        if args.translation_on:
            crop_seq = (args.st - 1, args.et)
            cropped.plot_translation(
                ax,
                location=crop_seq,
                fontdict={"size": 8, "weight": "bold"},
                y_offset=1,
            )
        ax.set_title(args.title)
        # Galaxy specific shenanigans
        tmp_fig = "./tmp.svg"
        plt.savefig(tmp_fig)
        plt.close()
    elif args.multiline:
        if args.sz:
            zoom_start, zoom_end = args.sz, args.ez
        else:
            zoom_start, zoom_end = 1, graphic_record.sequence_length
        cropped = graphic_record.crop((zoom_start, zoom_end))
        ax, _ = cropped.plot_on_multiple_lines(
            figure_width=args.plot_width,
            annotate_inline=lab_algo,
            figure_height=None,
            nucl_per_line=args.nucl_per_line,
            plot_sequence=False,
        )
        # ax.set_title(args.title)
        tmp_fig = "./tmp.svg"
        plt.savefig(tmp_fig)
        plt.close()
    else:
        ax, _ = graphic_record.plot(
            figure_width=args.plot_width, annotate_inline=lab_algo
        )
        ax.set_title(args.title)
        tmp_fig = "./tmp.svg"
        # Galaxy specific shenanigans
        plt.savefig(tmp_fig)
        plt.close()
    with open("tmp.svg", "rb") as img:
        for line in img:
            args.out_img.write(line)