You may be familiar with the Microsoft Biology Foundation (MBF), an open source bioinformatics .NET library that is designed to provide core functionality for genomics research from commonly-used file parsers to algorithms and web connectors. We’re pleased to announce that we are continuing to invest in the project in the coming year.
We recently posted a preview of MBF v2.0, which can be downloaded for development evaluation purposes. (Please note that this release is limited to source code only.) This release resolves a number of previously reported and identified issues. We also implemented-and in some cases, improved-many new features, including:
- BAM extension scenarios and paired end support
- Object model changes to use ISequence : IList<byte>
We also optimized some features for MBF v2.0, including:
- Memory profiling and analysis on MBF
- Parallel de Novo Assembler (PaDeNA) memory optimizations
- Sequence optimizations (including non-string and non-character)
- MUMmer optimization based on sequence
- Object model optimizations
- Additional scenarios for the collection of memory and performance profiles
We are now in the process of implementing additional features. We will discuss those in greater detail at a future date. In the meantime, we invite you to participate in the MBF v2.0 development preview and provide feedback through our community forums.
For best performance and to avoid any conflicts, we strongly recommend that you completely uninstall MBF v1.0 before you install MBF v2.0. Also, please note that we have identified the following issues that may affect your ability to install the code:
- The DLL version of MBF is incorrectly reported as 18.104.22.168
- The installer installs MBF in a directory named 2.1; it should be 2.0
The recommended (stable) MBF v1.0 code and corresponding v1.0 binaries are still available for download if you prefer to work with the earlier code. Either way, we encourage you to help us improve MBF by providing feedback in the forums.
For more information:
MBF download locations:
- Microsoft Biology Foundation v.2 (source code only)
- Microsoft Biology Foundation v.1 (recommended)
- Corresponding v.1 Binaries
-Beatriz Diaz Acosta, Senior Research Program Manager, Microsoft Research