.. _gettingstarted: Getting started =============== The functions that help analyzing data in the NMRlipids databank are described here. These functions are located in the :mod:`DatabankLib` module. To get started using these functions, first set up the package and initialize the databank: .. code-block:: bash pip install nmrlipids_databank nml_initialize_data toy source databank_env.rc "Toy" is a small test databank which is useful for testing and learning the package. You can then start to work with the `template `_ or write a code from the scratch. The minimum python code to intialize NMRlipids databank is .. code-block:: python import DatabankLib from DatabankLib.core import initialize_databank systems = initialize_databank() After running this, ``systems`` is an instance of :class:`DatabankLib.core.SystemsCollection` which works like a list but with added functionality and contains dictionaries where each dictionary is a simulation in the NMRlipids databank. A simulation dictionary contains the content of the README.yaml for that simulation. The content of README.yaml files is described in :ref:`Simulation metadata `. ``systems`` can be then used to loop over all simulations: .. code-block:: python for system in systems: print(system) Examples on analyses over NMRlipids databank can be found from the `template `_ and `codes used to analyze the results for the NMRlipids databank manuscript `_. The core functionality of the DatabankLib including initializing the databank and the definition of main data models is provided by :mod:`DatabankLib.core`. Some useful functions for analysis of the simulations can be found in the :mod:`DatabankLib.databankLibrary` module. .. toctree:: :maxdepth: 1 api/exampleAndTutorials