run_diffacto.py 文件源码

def zero_center_normalize(df, samples, logInput=False, method='median'):
    '''
    Transforming input peptide abundance table into log2-scale and centralize to zero.
    Inputs:
        df :        dataframe of peptide abundaces
        samples:    column names of selected samples
        logInput:   input abundances are already in log scale
        method:     method for estimating zero point
    '''
    assert method in ('median', 'average', 'GMM'), \
        'Zero centering method has to be among median, average or GMM!'
    if not logInput:
        # convert abundances to log2 scale
        df[samples] = df[samples].apply(np.log2)
    if method == 'average':
        norm_scale = np.nanmean(df[samples], axis=0)
    elif method == 'median':
        norm_scale = np.nanmedian(df[samples], axis=0)
    elif method == 'GMM':
        ''' two-component Gaussian mixture model '''
        from sklearn.mixture import GMM
        gmm = GMM(2)
        norm_scale = []
        for sp in samples:
            v = df[sp].values
            v = v[np.logical_not(np.isnan(v))]
            v = v[np.logical_not(np.isinf(v))]
            try:
                gmm.fit(np.matrix(v.values).T)
                vmean = gmm.means_[np.argmin(gmm.covars_)]
                norm_scale.append(vmean)
            except:
                norm_scale.append(np.nanmean(v))
        norm_scale = np.array(norm_scale)
    df[samples] = df[samples] - norm_scale
    return df