{
    "info": {
        "author": "Federico Tomasi",
        "author_email": "federico.tomasi@dibris.unige.it",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 4 - Beta",
            "Environment :: Console",
            "Intended Audience :: Developers",
            "Intended Audience :: Science/Research",
            "License :: OSI Approved :: BSD License",
            "Operating System :: MacOS",
            "Operating System :: POSIX",
            "Operating System :: Unix",
            "Programming Language :: Python",
            "Topic :: Scientific/Engineering :: Bio-Informatics",
            "Topic :: Software Development"
        ],
        "description": "# MIDAS\n\nThis repository contains python code that implements various estimators of entropy and mutual information.\nIn particular, Renyi Multi Information and Tsallis Multi Information estimators are available.\n\n## Quickstart\nInstall `midas` from the Python Package Index by using `pip`, via\n```bash\npip install midasML\n```\n\nAlternatively, you can clone it from our Github repository with\n\n```bash\n$ git clone https://github.com/SheffieldML/midas\n```\nand then install it with\n\n```bash\n$ python setup.py install\n```\nfrom the main folder of `midas`.\n\n## Working example\n\nTo test `midas`, you can use the following example.\nCreate a toy dataset specifying the correlation between some variables.\n```python\nimport numpy as np\nfrom midas.estimator import RenyiMutualInformationDivergence, MIDAS\nfrom midas.model_assessment import permutation_test_score_groups\nfrom midas.utils import sample_generation\n\nnp.random.seed(30)\nn_samples, n_dim = 30, 20\nrho_s = [.95, .9, .8, 0, 0]\nX, y, feature_names, groups = sample_generation.make_groups_joint(n_samples, n_dim, rho_s=rho_s)\n\nestimator = RenyiMutualInformationDivergence(alpha=0.99, k=3, n_iter=20)\nresult = permutation_test_score_groups(MIDAS(estimator), X, y, groups, n_jobs=-1)\n```\nThe first 3 groups of features are differentially co-regulated in half of the samples with respect to the other half, while the other groups of features are not (i.e., their co-regulation is the same in the two cases.)\nHence, we can use the `RenyiMutualInformationDivergence` to analyse the co-regulation for different classes of samples, by using `MIDAS` class.\n\n`result` is a `pandas.DataFrame` which contains a summary of the result, which is something like\n\n| | score        | perm_scores           | p-value  | group |\n|-----| ------------- |:-------------:| -----:|:-----|\n0 |\t0.614204 |\t[0.119599699041, 0.17002017399, 0.000642361597...|\t0.009901|\t[0, 1, 2, 3]\n1\t| 1.177062\t|[0.00106705500132, 0.0115119749325, 0.00023392...|\t0.009901\t|[4, 5, 6, 7]\n2\t| 0.223355\t|[0.0, 0.0410519694311, 0.00417966412452, 0.007...|\t0.029703\t|[8, 9, 10, 11]\n3\t| 0.010378\t|[0.00187253953604, 0.00757194385644, 0.0002593...|\t0.445545\t|[12, 13, 14, 15]\n4\t| 0.000000\t|[0.00307271647955, 0.0, 0.0129193052203, 0.010...|\t1.000000\t|[16, 17, 18, 19]\n\nThe first three results should have an high score and a low p-value, since the first three groups are differentially co-regulated by design (as specified before through the `rho_s` array). Hence, as in the example above, the method is capable to correctly address groups of variables as differentially co-regulated in the two classes of samples.\n\n## Other examples\nFor other examples, please refer to Jupyter notebooks present in our Github page at\nhttps://github.com/SheffieldML/midas/tree/master/notebooks.\n\n\n",
        "description_content_type": "",
        "docs_url": null,
        "download_url": "https://github.com/SheffieldML/midas/tarball/0.1.0",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://github.com/SheffieldML/midas",
        "keywords": "",
        "license": "FreeBSD",
        "maintainer": "",
        "maintainer_email": "",
        "name": "midasML",
        "package_url": "https://pypi.org/project/midasML/",
        "platform": "",
        "project_url": "https://pypi.org/project/midasML/",
        "project_urls": {
            "Download": "https://github.com/SheffieldML/midas/tarball/0.1.0",
            "Homepage": "https://github.com/SheffieldML/midas"
        },
        "release_url": "https://pypi.org/project/midasML/0.1.0/",
        "requires_dist": [
            "numpy (>=1.10.1)",
            "scipy (>=0.16.1)",
            "scikit-learn (>=0.18)"
        ],
        "requires_python": "",
        "summary": "Multi Information estimator for Differential Co-regulation Analysis",
        "version": "0.1.0"
    },
    "last_serial": 3690931,
    "releases": {
        "0.1.0": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "2c1d3e2ccdb22a7d96fa15def23b2ff1",
                    "sha256": "6fe8beb75aafdcdba45aa14e5a8197b8fca1d6888ed20d5bda9a7b34c3b4ca74"
                },
                "downloads": -1,
                "filename": "midasML-0.1.0-py2.py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "2c1d3e2ccdb22a7d96fa15def23b2ff1",
                "packagetype": "bdist_wheel",
                "python_version": "py2.py3",
                "requires_python": null,
                "size": 23423,
                "upload_time": "2018-03-21T10:21:55",
                "url": "https://files.pythonhosted.org/packages/d4/f7/94175a7cf6bcb6a610122ffb9a3c59d993151ba77704e4e55ce583270ba8/midasML-0.1.0-py2.py3-none-any.whl"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "2c1d3e2ccdb22a7d96fa15def23b2ff1",
                "sha256": "6fe8beb75aafdcdba45aa14e5a8197b8fca1d6888ed20d5bda9a7b34c3b4ca74"
            },
            "downloads": -1,
            "filename": "midasML-0.1.0-py2.py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "2c1d3e2ccdb22a7d96fa15def23b2ff1",
            "packagetype": "bdist_wheel",
            "python_version": "py2.py3",
            "requires_python": null,
            "size": 23423,
            "upload_time": "2018-03-21T10:21:55",
            "url": "https://files.pythonhosted.org/packages/d4/f7/94175a7cf6bcb6a610122ffb9a3c59d993151ba77704e4e55ce583270ba8/midasML-0.1.0-py2.py3-none-any.whl"
        }
    ]
}