To install nanite, use one of the following methods (the package dependencies will be installed automatically):
[CLI] makes sure that all dependencies for the
command line interface are installed. If you are
only using nanite as a Python module, you may safely omit it.
Note that if you are installing from source or if no binary wheel is available for your platform and Python version, Cython will be installed to build the required nanite extensions. If this process fails, please request a binary wheel for your platform (e.g. Windows 64bit) and Python version (e.g. 3.6) by creating a new issue.
What is nanite?¶
The development of nanite was motivated by a unique problem that arises in AFM force-distance data analysis, particularly for biological samples: The data quality varies a lot due to biological variation and due to experimental complexities that have to be dealt with when measuring biological samples. To address this problem, nanite makes use of machine-learning (á la scikit-learn), which allows to automatically determine the quality of a force-distance curve based on a user-defined rating scheme (see Rating workflow for more information). But nanite is much more than just that. It comes with an extensive set of tools for AFM force-distance data analysis.
Supported file formats¶
If you are a frequent AFM user, you might have run into several problems involving data analysis, ranging from simple data fitting to the visualization of quantitative force-distance maps. Here are a few usage examples of nanite:
- You would like to automate your data analysis pipeline from loading force-distance data to displaying a fit to the approach part with a Hertz model for a spherical indenter. You can do so with nanite, either via scripting or via the command-line interface that comes with nanite. For more information, see Fitting guide.
- You would like to automatically analyze and visualize maps of
force-distance data. This is possible with the
- You would like to sort force-distance data according to data quality using your own training set (not the one shipped with nanite). Nanite allows you to create your own training set from your own experimental data, locally. Besides that, you can make use of multiple regressors and visualize the rating e.g. of force-distance maps. For an overview, see Rating workflow.
If you are not interested in scripting, please have a look at the fitting guide.
In a Python script, you may use nanite as follows:
In : import nanite In : group = nanite.load_group("data/force-save-example.jpk-force") In : idnt = group # This group actually as only one indentation curve. In : idnt.apply_preprocessing(["compute_tip_position", ...: "correct_force_offset", ...: "correct_tip_offset"]) ...: In : idnt.fit_model(model_key="sneddon_spher") In : idnt.rate_quality() # 0 means bad, 10 means good quality Out: 9.060746150910978
You can find more examples in the examples section.