Created while conducting research for project on "Identifying and Ranking Amygdala Genes Linked to Autism Spectrum Disorder" at Children's National Health System

filtered.py

from collections import defaultdict
import numpy as np
import csv
np.set_printoptions(threshold=np.nan)

data = open("Data/GPL5175-3188.txt")
# Data for genes and associated IDs
geneList = data.read().split("\n")
data.close()
geneDict = {}
for i in range(13, 316932):
    geneDict[geneList[i].split("\t")[0]] = geneList[i].split("\t")[9]

data = open("Data/GSE25219-GPL5175_series_matrix.txt")
# Detailed data for each gene
geneData = data.read().split("\n")
data.close()

SFARIlist = defaultdict(list)
# Import SFARI gene list
data = open("Data/SFARI.csv")
SFARI = data.read().split("\n")
data.close()
for i in range(1, 281):
    dataRow = SFARI[i].split('\t')
    SFARIlist[dataRow[0].split(',')[1]].append(dataRow[0].split(',')[(len(dataRow)-4)])
#print(SFARIlist)
#print(len(SFARIlist))

SFARIset = set(SFARIlist)
#print(len(SFARIset))

SFARIKang = defaultdict(list)
for i in SFARIlist:
    for key, value in geneDict.items():
        # check if value contains substring
        if i in value:
            SFARIKang[i].append(key)
#print(SFARIKang)
print(len(SFARIKang))

Walshlist = []
# Import Walsh amygdala gene list
data = open("Data/Walsh Proband Amygdala Genes.csv")
Walsh = data.read().split("\n")
data.close()
for i in range(1, 341):
    dataRow = Walsh[i].split('\t')
    Walshlist.append(dataRow[0].split(',')[4])
#print(len(Walshlist))

Walshset = set(Walshlist)
#print(len(Walshset))

WalshKang = defaultdict(list)
for i in Walshlist:
    for key, value in geneDict.items():
        # check if value contains substring
        if i in value:
            WalshKang[i].append(key)
#print(WalshKang)
print(len(WalshKang))

SFARIWalshKang = {**SFARIKang, **WalshKang}
#print(SFARIWalshKang)
print(len(SFARIWalshKang))

# Make matrix of samples with relavant information
sample_titles = geneData[67].replace('"', '').split('\t')
sample_geo_accessions = geneData[61].replace('"', '').split('\t')
sample_organism = geneData[68].replace('"', '').split('\t')
sample_braincode = geneData[69].replace(
    '"', '').replace('brain code: ', '').split('\t')
sample_region = geneData[70].replace(
    '"', '').replace('region: ', '').split('\t')
sample_hemisphere = geneData[71].replace(
    '"', '').replace('hemisphere: ', '').split('\t')
sample_sex = geneData[72].replace('"', '').replace('Sex: ', '').split('\t')
sample_age = geneData[73].replace('"', '').replace('age: ', '').split('\t')
sample_stage = geneData[74].replace('"', '').replace('Stage: ', '').split('\t')
sample_postmortem_interval = geneData[75].replace(
    '"', '').replace('postmortem interval: ', '').split('\t')
sample_ph = geneData[76].replace('"', '').replace('ph: ', '').split('\t')
sample_rna_integrity = geneData[77].replace('"', '').replace(
    'rna integrity number: ', '').split('\t')
samples = np.column_stack((sample_titles, sample_geo_accessions, sample_organism, sample_braincode, sample_region,
                           sample_hemisphere, sample_sex, sample_age, sample_stage, sample_postmortem_interval, sample_ph, sample_rna_integrity))
# print(samples)

results = []
rows = []
csv = open('Gene List.csv', "w")
headers = "Gene Name, SFARI Rank [S-3], Period 2, Period 3, Period 4, Period 5, Period 6, Period 7, Period 8, Average Expression, Ratio, Walsh \n"
csv.write(headers)
for i in range(107, 17672): #17672
    Gene = key
    dataRow = geneData[i].replace('"', '').split('\t')
    total = []
    P2total = []
    P3total = []
    P4total = []
    P5total = []
    P6total = []
    P7total = []
    P8total = []
    for x in range(1, len(dataRow)):
        total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 2)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P2total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 3)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P3total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 4)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P4total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 5)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P5total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 6)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P6total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 7)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P7total.append(float(dataRow[x]))
        if(samples[x][4] == "AMY" and (int(samples[x][8]) == 8)):
            if(float(dataRow[x]) >= 6.0):
                results.append(dataRow[0])
            P8total.append(float(dataRow[x]))
    P2 = 0.0
    P3 = 0.0
    P4 = 0.0
    P5 = 0.0
    P6 = 0.0
    P7 = 0.0
    P8 = 0.0
    if(len(P2total) != 0):
        P2 = np.sum(P2total)/len(P2total)
    if(len(P3total) != 0):
        P3 = np.sum(P3total)/len(P3total)
    if(len(P4total) != 0):
        P4 = np.sum(P4total)/len(P4total)
    if(len(P5total) != 0):
        P5 = np.sum(P5total)/len(P5total)
    if(len(P6total) != 0):
        P6 = np.sum(P6total)/len(P6total)
    if(len(P7total) != 0):
        P7 = np.sum(P7total)/len(P7total)
    if(len(P8total) != 0):
        P8 = np.sum(P8total)/len(P8total)
    amyAvg = (P2 + P3 + P4 + P5 + P6 + P7 + P8)/7.0
    totalAvg = np.sum(total)/len(total)
    Ratio = amyAvg/totalAvg
    if(dataRow[0] in results):
        rowTemp = [dataRow[0], P2, P3, P4, P5, P6, P7, P8, amyAvg, Ratio]
        rows.append(rowTemp)
print(len(rows))

GenesFiltered = defaultdict(list)
for i in rows:
    for key, value in SFARIWalshKang.items():
        if(i[0] in value):
            GenesFiltered[key].append(i)
#print(GenesFiltered)
#print(len(GenesFiltered))

for key, value in GenesFiltered.items():
    SFARI = SFARIlist.get(key, "No")[0]
    Walsh = "No"
    if(key in Walshlist):
        Walsh = "Yes"
    row = key + "," + SFARI + "," + str(round((value[0])[1], 3)) + "," + str(round((value[0])[2], 3)) + "," + str(round((value[0])[3], 3)) + "," + str(round((value[0])[4], 3)) + "," + str(round((value[0])[5], 3)) + "," + str(round((value[0])[6], 3)) + "," + str(round((value[0])[7], 3)) + "," + str(round((value[0])[8], 3)) + "," + str(round((value[0])[9], 3)) + "," + Walsh + "\n"
    csv.write(row)