{
    "info": {
        "author": "Trevor Manz",
        "author_email": "trevor.j.manz@gmail.com",
        "bugtrack_url": null,
        "classifiers": [
            "Development Status :: 1 - Planning",
            "Intended Audience :: Science/Research",
            "Programming Language :: Python"
        ],
        "description": "# deep_lincs\n\nA deep learning wrapper around Keras for [Lincs](http://www.lincsproject.org/) L1000 expression data.\n\nCheck out the documentation [here](https://deep-lincs.readthedocs.io/en/latest/).\n\n## Installation\n```bash\n$ pip install deep-lincs\n```\n\n## Getting started\nThe data for 1.3 million L1000 profiles are availabe [on GEO](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE92742). The script load_files.sh fetches the `Level 3` data along with all metadata available. The largest file is quite big (~50Gb) so please be patient.\n\n```bash\n$ git clone https://github.com/manzt/deep_lincs.git && cd deep_lincs\n$ source load_files.sh # download raw data from GEO\n$ cd notebooks \n$ jupyter lab # get started in a notebook \n```\n\n## L1000 Dataset\nThe `Dataset` class is built with a variety of methods to load, subset, filter, and combine expression and metadata. \n```python\nfrom deep_lincs import Dataset\n\n# Select samples \ncell_ids = [\"VCAP\", \"MCF7\", \"PC3\"]\npert_types = [\"trt_cp\", \"ctl_vehicle\", \"ctl_untrt\"]\n\n# Loading a Dataset\ndataset = Dataset.from_yaml(\"settings.yaml\", cell_id=cell_ids, pert_type=pert_types)\n\n# Normalizing the expression data\ndataset.normalize_by_gene(\"standard_scale\")\n\n# Chainable methods\nsubset = dataset.sample_rows(5000).filter_rows(pert_id=[\"ctl_vehicle\", \"ctl_untrt\"])\n```\n\n## Models\nModels interface with the `Dataset` class to make training and evaluating different arcitectures simple.\n\n\n### Single Classifier\n\n```python\nfrom deep_lincs.models import SingleClassifier\n\nmodel = SingleClassifier(dataset, target=\"cell_id\")\nmodel.prepare_tf_datasets(batch_size=64)\nmodel.compile_model([128, 128, 64, 32], dropout_rate=0.1)\nmodel.fit(epochs=10)\n\nmodel.evaluate() # Evaluates on isntance test Dataset\nmodel.evaluate(subset) # Evalutates model on user-defined Dataset\n```\n\n### Multiple Classifier\n\n```python\nfrom deep_lincs.models import MutliClassifier\n\ntargets = [\"cell_id\", \"pert_type\"]\nmodel = MutliClassifier(dataset, target=targets)\nmodel.prepare_tf_datasets(batch_size=64)\nmodel.compile_model(hidden_layers=[128, 128, 64, 32])\nmodel.fit(epochs=10)\n\nmodel.evaluate() # Evaluates on isntance test Dataset\nmodel.evaluate(subset) # Evalutates model on user-defined Dataset\n```\n\n### Autoencoder\n\n```python\nfrom deep_lincs.models import AutoEncoder\n\nmodel = AutoEncoder(dataset)\nmodel.prepare_tf_datasets(batch_size=64)\nmodel.compile_model(hidden_layers=[128, 32, 128], l1_reg=0.01)\nmodel.fit(epochs=10)\n\nmodel.encoder.predict() # Gives encodings for instance test Dataset\nmodel.encoder.predict(subset) # Gives encodings for user-defined Dataset\n```\n\n\n",
        "description_content_type": "text/markdown",
        "docs_url": null,
        "download_url": "https://github.com/manzt/deep_lincs",
        "downloads": {
            "last_day": -1,
            "last_month": -1,
            "last_week": -1
        },
        "home_page": "https://deep-lincs.readthedocs.io",
        "keywords": "LINCS L1000 gene expression",
        "license": "MIT",
        "maintainer": "",
        "maintainer_email": "",
        "name": "deep-lincs",
        "package_url": "https://pypi.org/project/deep-lincs/",
        "platform": "",
        "project_url": "https://pypi.org/project/deep-lincs/",
        "project_urls": {
            "Download": "https://github.com/manzt/deep_lincs",
            "Homepage": "https://deep-lincs.readthedocs.io"
        },
        "release_url": "https://pypi.org/project/deep-lincs/0.0.3/",
        "requires_dist": null,
        "requires_python": "",
        "summary": "A framework for deep learning with LINCS L1000 Data",
        "version": "0.0.3"
    },
    "last_serial": 5659202,
    "releases": {
        "0.0.1": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "3335f06a34c8a466ba19c9550ed9b694",
                    "sha256": "ad75495be460b2a545cc732e67c94282f81df942b8cfda96585c73f0c8cb4a8b"
                },
                "downloads": -1,
                "filename": "deep_lincs-0.0.1.tar.gz",
                "has_sig": false,
                "md5_digest": "3335f06a34c8a466ba19c9550ed9b694",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 3713,
                "upload_time": "2019-08-03T16:11:58",
                "url": "https://files.pythonhosted.org/packages/e2/13/338f29ee06ad1c72a381ec22d4e8b74c5e55ae926cce82e292b9affd2b5a/deep_lincs-0.0.1.tar.gz"
            }
        ],
        "0.0.2": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "778063a9f0a0e96af1561f5a9e6834b4",
                    "sha256": "46d09583316f4d7ede4da7b93e40d77769a6638212412d45aad5bebcbebf43b9"
                },
                "downloads": -1,
                "filename": "deep_lincs-0.0.2.tar.gz",
                "has_sig": false,
                "md5_digest": "778063a9f0a0e96af1561f5a9e6834b4",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 20495,
                "upload_time": "2019-08-03T21:46:33",
                "url": "https://files.pythonhosted.org/packages/62/ce/5fb140b8c658b13409d1dcb9991f9287556ca6231d97916f52741430bdda/deep_lincs-0.0.2.tar.gz"
            }
        ],
        "0.0.3": [
            {
                "comment_text": "",
                "digests": {
                    "md5": "fde2c2ffaf6c167a2c762679d841a7c8",
                    "sha256": "36dd7e6a70d123b8e548d416de30825643c21565e7b277419d8609e4331a086b"
                },
                "downloads": -1,
                "filename": "deep_lincs-0.0.3-py3-none-any.whl",
                "has_sig": false,
                "md5_digest": "fde2c2ffaf6c167a2c762679d841a7c8",
                "packagetype": "bdist_wheel",
                "python_version": "py3",
                "requires_python": null,
                "size": 30237,
                "upload_time": "2019-08-10T10:38:36",
                "url": "https://files.pythonhosted.org/packages/44/dd/718eb6a672a95651c7b228cbdeea4c63aaa7ad8dafbb227c08590bb8aed6/deep_lincs-0.0.3-py3-none-any.whl"
            },
            {
                "comment_text": "",
                "digests": {
                    "md5": "e5e37fadc383410d299a33b7e841ff2e",
                    "sha256": "bcaed96605c8fc42a5e6112b648bb3152f4e9fa007a4079f4376311f4f48de51"
                },
                "downloads": -1,
                "filename": "deep_lincs-0.0.3.tar.gz",
                "has_sig": false,
                "md5_digest": "e5e37fadc383410d299a33b7e841ff2e",
                "packagetype": "sdist",
                "python_version": "source",
                "requires_python": null,
                "size": 22222,
                "upload_time": "2019-08-10T10:38:37",
                "url": "https://files.pythonhosted.org/packages/6c/08/46d796308fd2e80c4d41b4c0806f2b504108306b87115222501166ec0ea6/deep_lincs-0.0.3.tar.gz"
            }
        ]
    },
    "urls": [
        {
            "comment_text": "",
            "digests": {
                "md5": "fde2c2ffaf6c167a2c762679d841a7c8",
                "sha256": "36dd7e6a70d123b8e548d416de30825643c21565e7b277419d8609e4331a086b"
            },
            "downloads": -1,
            "filename": "deep_lincs-0.0.3-py3-none-any.whl",
            "has_sig": false,
            "md5_digest": "fde2c2ffaf6c167a2c762679d841a7c8",
            "packagetype": "bdist_wheel",
            "python_version": "py3",
            "requires_python": null,
            "size": 30237,
            "upload_time": "2019-08-10T10:38:36",
            "url": "https://files.pythonhosted.org/packages/44/dd/718eb6a672a95651c7b228cbdeea4c63aaa7ad8dafbb227c08590bb8aed6/deep_lincs-0.0.3-py3-none-any.whl"
        },
        {
            "comment_text": "",
            "digests": {
                "md5": "e5e37fadc383410d299a33b7e841ff2e",
                "sha256": "bcaed96605c8fc42a5e6112b648bb3152f4e9fa007a4079f4376311f4f48de51"
            },
            "downloads": -1,
            "filename": "deep_lincs-0.0.3.tar.gz",
            "has_sig": false,
            "md5_digest": "e5e37fadc383410d299a33b7e841ff2e",
            "packagetype": "sdist",
            "python_version": "source",
            "requires_python": null,
            "size": 22222,
            "upload_time": "2019-08-10T10:38:37",
            "url": "https://files.pythonhosted.org/packages/6c/08/46d796308fd2e80c4d41b4c0806f2b504108306b87115222501166ec0ea6/deep_lincs-0.0.3.tar.gz"
        }
    ]
}