Analysis of QY2H Data

https://github.com/dcluet/qY2H-Affinity-Ladder

Introduction

This program permits the automated generation of affinity ladders from quantitative Yeast Two Hybrid experiments. The program requires flow cytometry data .fcs files (linear scale). It generates a .csv table file that contains for each sample the mean reporter level. The actual affinity ladder graph is reported into a .pdf file.

For more information, the reader is referred to our article:

A quantitative tri-fluorescent yeast two-hybrid system: from flow cytometry to in-cellula affinities

Authors

CLUET David	david.cluet@ens-lyon.fr
SPICHTY Martin	martin.spichty@ens-lyon.fr

License

Copyright CNRS 2013

This software is a computer program whose purpose is to automatically analyze QY2H data (.fcs files) and generate in cellulo affinity ladder.

This software is governed by the CeCILL license under French law and abiding by the rules of distribution of free software. You can use, modify and/ or redistribute the software under the terms of the CeCILL license as circulated by CEA, CNRS and INRIA at the following URL: http://www.cecill.info/index.en.html

As a counterpart to the access to the source code and rights to copy, modify and redistribute granted by the license, users are provided only with a limited warranty and the software's author,the holder of the economic rights, and the successive licensors have only limited liability.

In this respect, the user's attention is drawn to the risks associated with loading, using, modifying and/or developing or reproducing the software by the user in light of its specific status of free software, that may mean that it is complicated to manipulate, and that also therefore means that it is reserved for developers and experienced professionals having in-depth computer knowledge. Users are therefore encouraged to load and test the software's suitability as regards their requirements in conditions enabling the security of their systems and/or data to be ensured and, more generally, to use and operate it in the same conditions as regards security.

The fact that you are presently reading this means that--delete the_remote_branch you have had knowledge of the CeCILL license and that you accept its terms.

Requirements

This program is optimized for Python 2.7 with the following libraries:

• datetime: To generate unique Analysis ID and file name.

• FlowCytometryTools v 0.4.6: To open .fcs files and manipulate flowcytometry data. http://eyurtsev.github.io/FlowCytometryTools/

• glob: To identify the .fcs files in the Input folder.

• matplotlib v 1.5.1: To generate the curves.

• numpy v 1.13.3: To generate and manipulate arrays.

• os: To handle paths of the raw data and generated files.

• Pillow / PIL v 3.1.2: To display images within the GUI of the program.

• sys: To permit manual abortion of the program.

• Tkinter v 8.6: To generate the GUI of the program.

Files

• README.md

• LICENSE.txt

• [] src

  • Analysis_QY2H.py

  • [] utils

    • __init__.py
    • channels.config
    • Colors.py
    • Configuration.py
    • Configure_Channels.py
    • Ending_Window.py
    • Functions.py
    • Logo.jpg
    • Object_Echantillon.py
    • Opening_window.py
    • Variables.py

• [] doc

  • Analysis_Configuration.jpg
  • Analysis_Progress.jpg
  • Logo_cnrs.jpg
  • Logo_ens.jpg
  • Logo_LBMC.jpg
  • Logo.jpg
  • Main_Menu.jpg
  • Results.jpg
  • Select_Channels.jpg
  • Select_File.jpg
  • Select_Input.jpg
  • Select_Output.jpg

User Guide

1 Recommendation for acquisition

Our program requires linear values for all fluorescence channels. Thus, be vigilant that your acquisition program is saving data as linear (even if your acquisition display is log or hyper log).

Yeast cells are usually smaller than the focused laser beam (spot) of flow cytometers. The maximum signal (= Height, H) for a given cell is obtained when the cell is fully covered by the laser spot. Thus, H reflects the total cellular content of the fluorophore. Therefore, we recommend to use the signal Height (H) of each channel.

Moreover, some flow-cytometers can apply internal corrections on specific channels. For example, the MacsquantVYB (that we used for our experiment) is correcting the Area A of each channels:

Area is the sum of a defined number of adjacent samples at the trigger time point divided by a scaling factor. This factor is chosen in a way that for "normal" events H=A to obtain a diagonal. The scaling factor is pressure dependent.

Thus we strongly recommend to use as much as possible non-manipulated values.

2 Main Menu

To start the program you need to execute the Analysis_QY2H.py python script:

$ python Analysis_QY2H.py

The main menu will propose you different functions:

1. Configure channels To select the channels to be used for the analysis.
2. Start analysis To generate a quantitative Yeast Two Hybrid affinity ladder from a set of .fcs files.
3. Abort To exit the program.

3 Configure the names of the channels

Before performing your first analysis, it is recommended to configure your channels. If you keep always the same acquisition settings, this step is required only once.

When clicking on Configure channels, the program prompts you to choose a .fcs file.

The program will identify all channels recorded in your file. You can then attribute the correct names in the various columns. For the subsequent analysis, the column BFP corresponds to the Reporter you want to quantify. The RFP and the GFP columns correspond to the BD-Bait and AD-Prey fusion proteins respectively.

The first two columns are not used yet, but might be included in a future development of this program to sub-select a population of cells with uniform FSC and/or SSC.

The select channel names will be saved in the channels.config file (in the utils folder) when clicking on VALIDATE.

4 Perform an analysis

When clicking on Start analysis, the program displays the analysis configuration window. You need first to select the folder where all your files (from the same experiment) are stored.

The program will automatically find all .fcs files present in this folder and display them in the analysis settings interface.

You need then to specific in which folder you want the output files to be generated. By default, the program is set on the input folder.

Once the path of the ÌNPUT and OUTPUT folders are set, you have access to the analysis settings. The program will generate the Affinity ladder by taking a sub-ensemble of cells using gates in the AD-Prey GFP and BD-Bait RFP channels. By default the minimal and maximal values are set to those of the Fig. 4 (B and C) of our publication.

The maximum in the Reporter (BFP) channel, corresponds to the upper-limit (x axis) of the generated Cumulative mean for each sample. If the curves in the .pdf output file are not reaching a plateau, increase this value.

The value BFP bins corresponds to the number of points you want to be displayed on the final graph.

You can remove the background of the system by selecting Remove negative Control. Unchecking this option is useful to monitor the contribution of the background in your experiment. This information is helpful especially for the weakest interactors.

As the sensitivity of the system may vary from one batch of yeast to an other, you can 'normalize' (to 100) the BFP signal using an Internal Reference. This will allow you to better compare various experiments. We recommend to use the strongest interaction as Internal Reference.

You need to specify which sample file corresponds to your negative control and Internal reference, even if no background subtraction or normalization are applied. Typically, the negative control corresponds to a qY2H experiment performed with fluorescent empty BD-Bait and AD-Prey fusion proteins. In our work, this control is called 0-0.fcs.

The value Number of cells corresponds to the maximum number of cells to be loaded from your file before doing the dual gating in the AD-Prey GFP and BD-Bait RFP channels. We highly recommend you to analyse at least 1 000 000 events to obtain a reliable affinity ladder.

You have the possibility to display the Cumulative mean in log or linear scale.

When clicking START, the program proceeds to the analysis (only if a negative control has been specified).

During the analysis the two Progress Bars inform you which file (first bar) is currently processed, and which analysis step (second bar) is performed.

Finally, the program displays the result of the analysis, with the main settings in the title. Here we present the result with the following activated options:

• Remove negative Control
• Normalize with the Internal Standard
• Y axis in linear scale

Click on ABORT to exit the program.

5 Output files

The program generates three files:

• A RESULT.png image of the graph presented at the end of the analysis
• A .csv table containing the mean BFP value for each sample file (after subtraction of the negative control and normalization, if selected)
• A .pdf report file, that encloses the qY2H affinity ladder graph.

The .csv and .pdf files have a common unique prefix based on the date and time of analysis. Moreover the data processing (i.e. background substraction and/or normalisation) is explicitly indicated.

6 Example files

The flow-cytometry files of qY2H experiments can be downloaded from http://flowrepository.org under accession numbers:

• FR-FCM-ZYUL (10 millions cells)
• FR-FCM-Z25G (1 million cells)