common.py 文件源码

def diff_exp(matrix, group1, group2, index):
    """Computes differential expression between group 1 and group 2
    for each column in the dataframe counts.

    Returns a dataframe of Z-scores and p-values."""

    g1 = matrix[group1, :]
    g2 = matrix[group2, :]

    g1mu = g1.mean(0)
    g2mu = g2.mean(0)

    mean_diff = np.asarray(g1mu - g2mu).flatten()
    # E[X^2] - (E[X])^2
    pooled_sd = np.sqrt(
        ((g1.power(2)).mean(0) - np.power(g1mu, 2)) / len(group1)
        + ((g2.power(2)).mean(0) - np.power(g2mu, 2)) / len(group2))
    pooled_sd = np.asarray(pooled_sd).flatten()

    z_scores = np.zeros_like(pooled_sd)
    nz = pooled_sd > 0
    z_scores[nz] = np.nan_to_num(mean_diff[nz] / pooled_sd[nz])

    # t-test
    p_vals = (1 - stats.norm.cdf(np.abs(z_scores))) * 2

    df = pd.DataFrame(OrderedDict([('z', z_scores), ('p', p_vals)]),
                      index=index)

    return df