phconvert is a python 2 & 3 library that helps writing valid Photon-HDF5 files, a file format for time stamp-based single-molecule spectroscopy. Additionally, phconvert can convert to Photon-HDF5 all the common binary formats used in solution-based single-molecule spectroscopy. These includes PicoQuant’s .HT3/.PT3/.PTU/.T3R, Becker & Hickl’s .SPC/.SET and the .SM format used by WeissLab and others for µs-ALEX smFRET.
Nov. 2018: Phconvert 0.9 released, see the release notes.
Quick-start: Converting files to Photon-HDF5
Converting one of the supported files formats to Photon-HDF5 does not require being able to program in python. All you need is running the “notebook” corresponding to the file format you want to convert from, and follow the instructions therein.
For demonstration purposes, we provide a demo service to run the notebooks online without any installation. With this online service, you can convert data files up to 35MB to Photon-HDF5. To launch the demo click on the following button (see also instructions):
To execute the phconvert notebooks on your machine, you need to install the Jupyter Notebook App first. A quick-start guide on installing and running the Jupyter Notebook App is available here:
Next, you need to install the phconvert library with the following command
(type it in Terminal on OS X or Linux, or in the
cmd prompt on Windows):
conda install -c conda-forge phconvert
Finally, you can download one of the provided notebooks and run it on your machine.
Simply, download the
which contains all the notebooks in the
For questions or issues:
If you have a file format that is not yet supported, please open an new Issue. We are willing add support for as many file formats as possible!
When writing Photon-HDF5 files, phconvert saves you time and protects you against common errors that risk to make the file not a valid Photon-HDF5. Also a description is automatically added to each Photon-HDF5 field. The descriptions are extracted from a JSON file which contains the list Photon-HDF5 field names, types, and descriptions.
See also Writing Photon-HDF5 files in the Photon-HDF5 reference documentation.
Read Photon-HDF5 files
In case you just want to read Photon-HDF5 files you don’t need to use phconvert. Photon-HDF5 files can be directly opened with a standard HDF5 viewer HDFView.
See also Reading Photon-HDF5 files in the Photon-HDF5 reference documentation.
The recommended way to install phconvert is using conda:
conda install -c conda-forge phconvert
If you don’t have conda installed, please install the free python distribution Anaconda choosing the python 3 version. Starting from version 0.9, the aging python 2.7 is not supported anymore.
Alternatively, you can install phconvert in any python installation using PIP:
pip install phconvert
In this latter case, make sure that numpy and pytables are installed.
- python 3.4 or greater (3.6+ recommended)
- numpy >=1.9
- pytables >=3.1
- numba (optional) for faster PicoQuant files decoding
Note when installing via
condaall the dependencies are automatically installed.
The phconvert library documentation (for developers)
The phconvert API documentation can be found on ReadTheDocs:
phconvert is released under the open source MIT license.
As with other Photon-HDF5 subprojects, we encourage contributions in any form, from simple suggestions, typo fix to the addition of new features. Please use GitHub by opening Issues or sending Pull Requests.
All the contributors will be acknowledged in this website, and will included as authors in the next software-paper publication.
For more details see our contribution policy.
Authors & Contributors
List of contributors:
- Antonino Ingargiola (@tritemio)
- Ted Laurence (@talaurence)
- Marco Lamperti (@lampo808)
- Xavier Michalet (@smXplorer)
- Anders Barth (@AndersBarth)
- Biswajit Pradhan (@biswajitSM) <biswajitp145 AT gmail.com.
- Sébastien Weber (@seb5g)
- David Palmer (@dmopalmer)
We thank also @ncodina for providing PTU files and helping in testing the PTU decoder in phconvert.
This work was supported by NIH Grant R01-GM95904.
Release 0.9 was supported by Prof. Eitan Lerner.