Ken-Kuroki

# Make sure you have downloaded ftp://ftp.ncbi.nih.gov/pub/taxonomy/taxdmp.zip
# and have its contents in ./data/.

# Note: This is just another reinvention of NCBITaxonomy of ETE Toolkit.
# http://etetoolkit.org/docs/latest/tutorial/tutorial_ncbitaxonomy.html

import pandas as pd
from collections import defaultdict

Ken-Kuroki

You may want to align sequences with Biopython, and after reading its tutorial ( http://biopython.org/DIST/docs/tutorial/Tutorial.html#htoc87 ), you will try

>>> from Bio import Align
>>> aligner = Align.PairwiseAligner()

only to encounter

AttributeError: module 'Bio.Align' has no attribute 'PairwiseAligner'

Ken-Kuroki

Ken-Kuroki

import doctest
import xml.etree.ElementTree as ET

from Bio import Entrez

Entrez.email = "example@example.com"  # replace it with your actual email address

def get_taxonomic_name(taxid, rank):
    '''Return the taxonomic name of the specified rank for the given taxonomy ID.
    Make sure you give a string, not an int value as taxid or efetch throws an exception.

Ken-Kuroki

import pandas as pd

def read_blast_output(output):
    """Reads BLAST output (outfmt 6) and returns a pandas dataframe."""
    
	return pd.read_csv(output,
                       sep="\t",
                       names=["qseqid", "sseqid", "pident", "length", "mismatch", "gapopen", "qstart", "qend", "sstart", "send", "evalue", "bitscore"],
                       index_col="qseqid")

Ken-Kuroki

#  For plotting with log-scale

def makebin(start, end):
    bins = []
    for i in range(start, end, 1):
        bins.append(10**i)
    return bins

Ken-Kuroki

What's this?

Given a NCBI feature table that lists coding sequences, just wanted to calculate length of non-coding regions with pandas. This is obviously easy when you iterate a dataframe, but that can be painfully slow. Instead, here's how to do it in a "pandas-like" way.

def calc_nc_len(df):
    # Before and after here merely means coordinates on the genome disregarding the coding strand of each gene
    df["before"] = df["end"].diff()
    df["before"] -= df.apply(lambda x: x["end"] - x["start"],axis=1)
    df["after"] = -df["start"].diff(periods=-1)
    df["after"] -= df.apply(lambda x: x["end"] - x["start"],axis=1)