FabLab

Integrated LC-MS/MS-based de novo sequencing of antibody chains in complex mixtures

Getting started

Download and run the latest installer (fablab_installer.exe) from the releases page. FabLab is only compatible with windows as the graphical user interface (GUI) was built using Windows Forms.

Building/Developing

To build FabLab from source, download the source code for FabLab using git clone.

git clone https://github.com/Bdegraaf1234/FabLab.git

The project is built using dotnet (.NET Framework 4.8) and development is done on Windows using visual studio. To build the project, install dotnet and enter the following commands into your terminal from the root folder of this repository, FabLabPublic/.

dotnet restore /p:configuration=release
dotnet build /p:configuration=release

This will generate an executable, FabLab.exe, in the build folder (default: ./FabLab/bin/release/net4.8/). The executable needs the other files in this folder to run so it is advised to either run the software from the build folder or generate an installer using for example inno-setup.

Examples

In the ./examples folder you can find several example datasets. If you have installed the software, you can click any of the .flconfig files and the software will start after preprocessing the data (this may up to 1 minute). If you have built the software yourself you can start it by running the FabLab.exe file. You can then click File --> Load --> From config file and navigate to one of the .flconfig files. You are encouraged to play around with these datasets before you move on to your own data, as the software is relatively unstable and requires some experience to operate. For a detailed elaboration on the testing data and underlying algorithms, you are referred to the publication in ./CITATION.cff.

Examples are organized per target chain, and include 1-3 .flconfig files. A good start would be the polyclonal_IgA1 example and opening the .flconfig files in the following order while following along with the selection process by exploring the plots and scores:

Examples/polyclonal_IgA1/1.framework_region_sequencing.flconfig
Examples/polyclonal_IgA1/2.cdr_sequencing.flconfig
Examples/polyclonal_IgA1/3.chain_sequencing.flconfig

The example datasets work fine with the default settings. Navigating the user interface is done with the mouse. Tables, column headers and rows can be right-clicked to reveal contextual options. On the left side of the main screen you can find several settings that you can change.

Analyzing your own data

All tables can be filtered and sorted by right-clicking the column headers. Filtered segment candidates should be selected for in depth scoring/inspection in a separate view (rightclick, Analyze --> Rescore all shown candidates for this region) and exported (rightclick, Write --> ...) for subsequent stages. Broadly, you should aim to reject incorrect FR candidates using the scores, graphs and annotated spectra. You can select framework region (FR) candidates for further analysis (Select for prediction to analyze directly, or Check contig for recombination for automated analysis of multiple combinations. When you have selected one or more adjacent FR pairs (e.g. one FR1 and three FR2) then you can recombine the into FR-CDR-FR candidates. This is initially done by rightclicking a CDR (Complementarity Determining Region) table, and clicking Analyze --> Recombine adjacent checked candidates. A new tab with scored FR-CDR-FR candidates will then open. Export the best of these candidates and create a new .flconfig file to move on to the next stage (or in the case of the examples, just open the next .flconfig file).

Preprocessing

de novo peptides

De novo peptides for analysis were generated from shotgun LC-MS/MS data using PEAKS de novo sequencing software. To support other sources of de novo peptides, open an issue or contact the authors. These peptides were analyzed using the STITCH Bottom-Up sequencing tool to select a germline template from the IMGT database and generate a starting pool of FR candidates (proforma format).

Middle-Down fragmentation data

Middle down data was acquired as described in the publication in ./CITATION.cff. These data were deconvoluted using FreeStyle and BioPharmaFinder. A potential (Free) alternative is ms_deisotope. Deconvoluted Middle down spectra should be supplied in mgf format, where all fragment ions have been converted to monoisotopic singly charged masses. The precursor mass is given as an average, singly charged mass.

Credits

Developers

• Bastiaan de Graaf - Software engineer - s.c.degraaf{at}uu{dot}nl
• Douwe Schulte - collaborating developer, lead developer for the Stitch Bottom-Up sequencing tool

Academic guidance

• Albert Heck - Principal investigator
• Richard Scheltema - Developer of the hecklib dependencies

Eperimental support

• Sem Tamara - Experimental method development
• Max Hoek - Experimental method development
• Albert Bondt - Experimental method development
• Weiwei Peng - Experimental method development

Acknowledgements

• All credited individuals are part of the group "Biomolecular Mass Spectrometry and Proteomics" (or here) at the Utrecht University

Dependencies

• FabLab is built on the Hecklib MS Data analysis packages. These packages are precompiled into a single nuget dependency for direct consumption, located in the nuget/ folder.

License

GPLV2 License (see LICENSE.md)