Mercurial > repos > jay > pdaug_sequence_similarity_network
view PDAUG_Word_Vector_Descriptor/PDAUG_Word_Vector_Descriptor.py @ 0:e650de82bcc7 draft
"planemo upload for repository https://github.com/jaidevjoshi83/pdaug commit a9bd83f6a1afa6338cb6e4358b63ebff5bed155e"
author | jay |
---|---|
date | Wed, 28 Oct 2020 01:50:00 +0000 |
parents | |
children | 5ae3966929db |
line wrap: on
line source
import numpy as np import os import pandas as pd from Bio import SeqIO from nltk import bigrams from nltk import trigrams import gensim import argparse parser = argparse.ArgumentParser() parser.add_argument("-M", "--ModelInput", required=True, default=None, help="Path to target tsv file") parser.add_argument("-R", "--row", required=True, default=None, help="Path to target tsv file") parser.add_argument("-I", "--InputFasta", required=True, default=6, help="Path to target tsv file") parser.add_argument("-O", "--OutFile", required=False, default='model.txt', help="Path to target tsv file") parser.add_argument("-P", "--positive", required=True, help="Path to target tsv file") parser.add_argument("-N", "--negative", required=True, help="Path to target tsv file") args = parser.parse_args() seed = 42 np.random.seed(seed) new_model = gensim.models.KeyedVectors.load_word2vec_format(args.ModelInput, binary=False) import time t0 = time.time() temp_word = np.zeros(shape=(int(args.row), 200)) for index, seqs in enumerate(SeqIO.parse(args.InputFasta, 'fasta')): seq_sum = 0 tri_seq = trigrams(seqs.seq) for item in ((tri_seq)): tri_str = item[0] + item[1] + item[2] if tri_str not in list(new_model.wv.vocab): continue seq_sum = seq_sum + new_model[tri_str] temp_word[index] = seq_sum t1 = time.time() temp_word = temp_word clm = [x for x in range(0,temp_word.shape[1])] y_temp_word = np.vstack((np.ones((int(args.positive), 1)), np.zeros((int(args.negative),1)))) c, r = y_temp_word.shape y_temp_word = y_temp_word.reshape(c,) class_label = pd.DataFrame(y_temp_word, columns=["Class_label"]) df = pd.DataFrame(temp_word, columns=clm) df = pd.concat([df,class_label], axis=1) df.to_csv(args.OutFile, index=None, sep="\t")