scores.py 文件源码

def getAabrhRawScoreSummmaryD(strainNamesL,aabrhL,scoresO,geneNames):
    '''Given raw scores and a directory with blast output, finds the sets of all around best reciprocal hits. Then for each pair of species, calculates the mean and standard deviation of scores and stores in a dictionary.'''

    # now loop through these, sorting scores into a dict keyed by species pair.

    # create dictionary, (representing an upper triangular matrix)
    spScoreD={}
    for i in range(len(strainNamesL)-1):
        strain1 = strainNamesL[i]
        for j in range(i+1,len(strainNamesL)):
            strain2 = strainNamesL[j]
            spScoreD[(strain1,strain2)]=[]

    # loop through aabrhL and populate
    for orthoT in aabrhL:
        spScoreD = addPairwiseScores(spScoreD,orthoT,scoresO,geneNames)

    # get mean and standard deviation
    summaryD = {}
    for sp1,sp2 in spScoreD:
        mean = statistics.mean(spScoreD[(sp1,sp2)])
        std = statistics.stdev(spScoreD[(sp1,sp2)])
        summaryD[(sp1,sp2)] = (mean,std)
        summaryD[(sp2,sp1)] = (mean,std)

    return summaryD