Transcriptome Analysis of Spermophilus lateralis and Spermophilus tridecemlineatus Liver Does Not Suggest the Presence of Spermophilus-Liver-Specific Reference Genes

Algorithm 2

Python codes for expression analysis.

Datafile  = ‘normalized_GSE2021.csv’

sample_output = ‘NF_correlation.csv’

random_output  = ‘random_correlation.csv’

sample_list  =  [‘02n12’, ‘13d19’, ‘05p15’, ‘02g16’, ‘03o23’,

              ‘03j13’, ‘13d08’, ‘28h07’, ‘13o12’, ‘12p20’]

import math

import random

def process(x, y):

 mean_x  =  float(sum(x)) / len(x)

 mean_y =  float(sum(y)) / len(y)

 covariance =  sum([((x[i]−  mean_x) * (y[i]−  mean_y))

            for i in range(len(x))])

 error_x =  [i  −  mean_x for i in x]

 error_y =  [i  −  mean_x for i in y]

 sd_x =  math.sqrt(sum([i*i for i in error_x]))

 sd_y =  math.sqrt(sum([i*i for i in error_y]))

 gradient =  sum([error_x[index] * error_y[index]

           for index in range(len(error_x))]) /  ∖

        sum([error_x[index] * error_x[index]

           for index in range(len(error_x))])

 intercept =  mean_y −  (gradient * mean_x)

 return (float(covariance), float(sd_x), float(sd_y),

     float(gradient), float(intercept))

fdata =  [x[:−1].split(‘,’) for x in open(datafile, ‘r’).readlines()]

# Sorting out the full data (fdata) into 3 parts

# sdata -  contains data for IDs in sample_list

# rdata -  contains data (n =number of IDs in sample_list)

#  for IDs not in sample_list

# data -  contains data for IDs not in sample_list or rdata

sdata =  [x for x in fdata if x[0] in sample_list]

data =  [x for x in fdata if x[0] not in sample_list]

rdata =  [random.choice(data) for x in range(len(sample_list))]

data =  [x for x in data if x[0] not in [r[0] for r in rdata]]

sout =  open(sample_output, ‘w’)

rout =  open(random_output, ‘w’)

print str(len(sdata)), ‘ number of samples in sample list’

print str(len(rdata)), ‘ number of samples in random list’

print str(len(data)), ‘ number of samples in data list’

sout.write(‘,’.join([‘sample_x’, ‘sample_y’, ‘covariance’,

 ‘sd_x’, ‘sd_y’, ‘gradient’, ‘intercept’]) + ‘∖n’)

rout.write(‘,’.join([‘sample_x’, ‘sample_y’, ‘covariance’,

 ‘sd_x’, ‘sd_y’, ‘gradient’, ‘intercept’]) + ‘∖n’)

def run_correlation(sample_data, other_data, outfile):

 count =  1

 for s in sample_data:

  ID1 =  s[0]

  d1 =  [float(x) for x in s[1:]]

  for x in other_data:

   ID2 =  x[0]

   d2 =  [float(y) for y in x[1:]]

   result =  process(d1, d2)

   result =  [ID1, ID2, str(result[0]), str(result[1]),

        str(result[2]), str(result[3]), str(result[4])]

   outfile.write(‘, ’.join(result) + ‘∖n’)

   count =  count + 1

  print str(count), ‘ processed. ID =’, ID1

 outfile.close()

print ‘Processing sample list’

run_correlation(sdata, data, sout)

Print

print ‘Processing random list’

run_correlation(rdata, data, rout)