IMGT/StatClonotype Documentation





http://www.imgt.org
 IMGT®, the international ImMunoGeneTics information system®

---

• Introduction
• 'IMGTStatClonotype' R package installation
• IMGT/StatClonotype web tool
  • Launching IMGT/StatClonotype web tool
  • Selection of the two sets to be compared
  • Display parameters selection
  • Results
    1. IMGT/HighV-QUEST set 1
    2. IMGT/HighV-QUEST set 2
    3. Statistical test results
          3.1. Statistical test results for genes
          3.2. Statistical test results for alleles
    4. Multiple testing procedures plots
          4.1. Multiple testing procedures plots for genes
          4.2. Multiple testing procedures plots for alleles
    5. Synthesis graph
          5.1. Synthesis graph for genes
          5.2. Synthesis graph for alleles
    6. CDR-IMGT lengths
    7. CDR-IMGT positions
    8. V-D-J gene associations
  • Exiting IMGT/StatClonotype
• IMGT/StatClonotype implementation and developments
• IMGT/StatClonotype upgrades and versions
• 'IMGTStatClonotype' R package update

• IMGT/StatClonotype [1], developed by LIGM (Montpellier University, CNRS) and part of IMGT®, the international ImMunoGeneTics information system® (http://www.imgt.org) [2] is a standalone IMGT® tool for statistical analysis of sets from IMGT/HighV-QUEST output [3,4], on the Web since June 2016.
  
  IMGT/StatClonotype works only with uploaded files (.txt format) from the IMGT/HighV-QUEST statistical analysis output ('stats_xxx' file available in 'data' directory of the IMGT/HighV-QUEST statistical analysis output where 'xxx' is the batch name and the locus type) as described in section B.
  To get these files, you therefore need to launch the IMGT/HighV-QUEST statistical analysis (IMGT/HighV-QUEST Documentation).
  
  
• IMGT/StatClonotype uses a generic statistical procedure [1] for identifying significant changes in IG and TR differences of proportions of IMGT clonotypes (AA) diversity and expression [5].
  It uses the IMGT gene and allele nomenclature [6,7,8] based on IMGT-ONTOLOGY [9] and IMGT standards in immunoinformatics [10].
  
  
• IMGT/StatClonotype is a tool incorporated in a new R package ('IMGTStatClonotype') to perform pairwise comparison of sets from IMGT/HighV-QUEST output through a user-friendly web interface implemented using Shiny framework [11] in users' own browser.

• To install 'IMGTStatClonotype' R package you first need to install the R program on your computer.
  R is a language and environment for statistical computing and graphics available as free software and downloadable at the CRAN (Comprehensive R Network) website (http://cran.r-project.org/).
  
  Note that: If R is already installed on your computer please check if you have a recent version (>=4.0.3).
  
  To install R for the first time please follow the steps below:
  
  For Windows operating system
• Step 1: Go to https://cran.r-project.org/ and click on "Download R for Windows" under Download and Install R.
• Step 2: Click on the link "base" under "Subdirectories".
• Step 3: Click on the link "Download R X.X.X for windows" where X.X.X gives the R version (e.g., R 4.0.3). An executable file (e.g. R-4.0.3-win.exe) will be saved on your computer. Click on the icon of that file to run it.
• Step 4: Choose language to install it. (e.g., English)
• Step 5: Click "Next" at the bottom of the "Information" window.
• Step 6: Select the following where R will be installed (e.g. "C:\Program Files") and click "Next" at the bottom of the "Select Destination Location" window.
• Step 7: Click "Next" at the bottom of the "Select Destination Location" window.
• Step 8: Click "Next" at the bottom of the "Select Components" window.
• Step 9: Click "Next" at the bottom of the "Startup options" window.
• Step 10: Click "Next" at the bottom of the "Select Additional Tasks" window.
• Step 11: Click "Finish" when the setup has finished installing R on your computer.
• Step 12: Two "R" icon appears on the desktop of the computer corresponding to 32 and 64 bits processors. Double-click on one of those icons to start R.
  You can also find the "R" icon by clicking on the button "Start" at the bottom left of your computer screen and choosing "All programs" then selecting Rx64 4.0.3 for example in the "R" folder.
  The R console should pop up:
  
  
  
  
  
  
• Step 13: To install the package 'IMGTStatClonotype' you first have to install dependencies packages (reshape2 (>= 1.4.1), data.table (>= 1.9.6), multtest (>=2.24.0), ggplot2 (>= 2.1.0), gridExtra (>=2.0.0), DT (>= 0.1), shiny (>= 0.13.2), shinyjs (>= 0.4.0), colourpicker(>=0.3), plotly (>=3.4.13), htmlwidgets (>=1.5.3), png (>=0.1-7), dendextend (>=1.14.0)) by typing into the R console each line of code (in red and comments in green) given below:
  
  
  install.packages(c("reshape2", "data.table", "ggplot2", "shiny", "shinyjs", "colourpicker", "gridExtra", "DT", "plotly", "htmlwidgets", "png", "dendextend")) # Choose the website you want to download the package from, you should choose "France" (or another country)
  wspace <- getwd()# Press ENTER
  dest =paste(wspace,"/d3heatmap_0.6.1.tar.gz",sep="")# Press ENTER
  download.file("http://www.imgt.org/download/StatClonotype/d3heatmap_0.6.1.tar.gz",destfile=dest)# Press ENTER
  install.packages("d3heatmap_0.6.1.tar.gz", repos=NULL)# Press ENTER
  if (!requireNamespace("BiocManager", quietly = TRUE)) install.packages("BiocManager")# Press ENTER
  BiocManager::install("multtest") # If a message appears to update a list of package (Update all/some/none? [a/s/n]) type n then press ENTER
  
  
  
• Step 14: 'IMGTStatClonotype' can be easy installed from the R console
  
  
  des=paste(wspace,"/IMGTStatClonotype_1.0.4.tar.gz",sep="")# Press ENTER
  download.file("http://www.imgt.org/download/StatClonotype/IMGTStatClonotype_1.0.4.tar.gz",destfile=des)# Press ENTER
  install.packages("IMGTStatClonotype_1.0.4.tar.gz", repos=NULL)# Press ENTER
  
  
  
  The 'IMGTStatClonotype' package is now installed.
  
  
  
  
  
  See the user manual for more details: IMGTStatClonotype-manual
  
  
  For non-Windows operating system (Linux or Macintosh)
  
  
• To install R on a non-Windows operating system you should go to https://cran.r-project.org/ and click on "Download R for Linux" or "Download R for (Mac) OS X" under 'Download and Install R' and follow the instructions at http://ftp.heanet.ie/mirrors/cran.r-project.org/doc/FAQ/R-FAQ.html.
  
  
  
  

A. Launching IMGT/StatClonotype

library(IMGTStatClonotype)# Press ENTER
launch() # Press ENTER

B. Selection of the two sets to be compared

• Two sets to be compared should be chosen into IMGT/StatClonotype among the sets you have uploaded from the IMGT/HighV-QUEST output ('stats_xxx' file available in 'data' directory of the IMGT/HighV-QUEST statistical output where 'xxx' is the batch name and the locus type (e.g., 'stats_MID1_326382_TRB')). The format of the file is described in IMGT/HighV-QUEST Documentation and in page 8 of the IMGTStatClonotype manual
  . Files examples are provided in MID1.txt and MID2.txt.
  
  
• To test the tool IMGT/StatClonotype you can download the files MID1.txt and MID2.txt at IMGT/StatClonotype and you can choose them as inputs for 'IMGT/HighV-QUEST set 1' and 'IMGT/HighV-QUEST set 2', respectively.
  
  

C. Display parameters selection

D. Results

1. IMGT/HighV-QUEST set 1

• The first tab 'IMGT/HighV-QUEST set 1' displays IMGT clonotypes (AA) with selected CDR3-IMGT lengths defined for the statistical procedure are displayed in tables at the top.
  
  
• Unselected IMGT clonotypes (AA) with outlier CDR3 lengths are also listed below.
  
  
• The following information is provided for each clonotype (AA): CDR3-IMGT sequence (AA), Experimental ID, Representative sequence index, Nb of '1 copy', Nb of 'More than one', Total nb of '1 copy' and 'More than one', '1 copy' Indexes, V-gene, V-allele, D-gene, D-allele, J-gene, J-allele, CDR1-IMGT, CDR2-IMGT, CDR1-IMGT gapped sequence (AA), CDR2-IMGT gapped sequence (AA), V-REGION %identity, Sequence length, C104, F118 or W118, Anchors, Sequence ID, Functionality, Sequence file number, sequence clonotype file number.

2. IMGT/HighV-QUEST set 2

3. Statistical test results

• The third tab 'Statistical test results' displays the results of the significance of the difference in proportions with 95% confidence intervals (CI) of IMGT/HighV-QUEST output for IMGT clonotype (AA) diversity or expression.

  3.1. Statistical test results for genes

  • The first table display test results for genes.
    
    
    
    
    
    
    
    - Gene name, Gene type, Nb of IMGT clonotypes (AA), Proportion and Normalized proportion for set 1 and for set 2, Difference in proportions, z-scores values, Lower and Upper bound confidence interval (CI) of the difference in proportions are given from column 1 to column 12 of the table.
    
    - Unadjusted p-values (rawp) and adjusted p-values from multiple testing procedures (Bonferroni, Holm, Hochberg, ŠidákSS, ŠidákSD, BH, BY) are given from column 13 to column 20 of the table.
    
    Reminder: Properties of the multiple testing procedures (see [1] for more details)
    
    

    Procedures	Type of control	Algorithm structure	dependence of p-values under H0	Properties
    Bonferroni	FWER	Single-step	Ignorance	The most conservative
    Šidák (SS)	FWER	Single-step	Independence	Less conservative than Bonferroni
    Holm	FWER	Step-down	Ignorance	Less conservative than Bonferroni
    Šidák (SD)	FWER	Step-down	Dependence	Similar to Holm
    Hochberg	FWER	Step-up	Independence	Step-up of Holm
    Benjamini & Hochberg (BH)	FDR	Step-up	Independence	The least conservative
    Benjamini & Yekutieli (BY)	FDR	Step-up	Ignorance	More conservative than BH

    
    We assume that the two compared sets are independent and the individual tests are independent of each other (i.e., multiple hypotheses are independent).
    
    - Significant unadjusted and adjusted p-values are highlighted in yellow and the interpretation of the test is given in the last column of the table.
    
    - Analysis for several or single alleles can be selected through the left panel.
    
    - The table of results is downloadable in a .csv format with possibility of selecting particular columns through the left panel (for the display in the web page the columns 'z', 'rawp', 'Bonferonni', 'Holm', 'Hochberg', 'ŠidákSS', 'ŠidákSD', 'BH' and 'BY' must be selected together in the left panel).
  
  
  • The second table displays the list of genes with null or small occurrences. Gene name, Gene type, Nb of IMGT clonotypes (AA), Proportion and Normalized proportion for set 1 and for set 2 are given in this table.
    
    For this list of genes the Fisher's exact test is used and associated p-values are displayed in "rawp" column of the "Test results for genes" table.
    In this case of null or small gene occurrences
    n_i p_i^k<5 , n_i (1-p_i^k)<5 and n_j p_j^k <5 , n_j (1-p_j^k)<5
    
    where {i,j}_(i≠j): set indexes, n_i (resp., n_j): number of IMGT clonotypes (AA) in set i (resp., set j) , k: IMGT gene name and p_i^k (resp., p_j^k): proportion of the gene k in set i (resp. set j), the z-score is not applicable and results displayed in "z" column are not considered.
    
    -Multiple testing procedures are applied for p-values returned by the Fisher's exact test.
    
    -Null or small occurrences figured in this table can be hidden through the left panel.
    
    

  3.2. Statistical test results for alleles

  • The third table display test results for alleles.
    
    Only alleles of genes having significant differences in proportions validated by all multiple testing procedures are analyzed.
    
    - Allele name, Allele type, Nb of IMGT clonotypes (AA), Proportion, Normalized proportion for set 1 and set 2, Difference in proportions, z-scores values, Lower and Upper bound confidence interval (CI) of the difference in proportions are given from column 1 to column 12 of the table.
    
    - Unadjusted p-values (rawp) and adjusted p-values from multiple testing procedures (Bonferroni, Holm, Hochberg, ŠidákSS, ŠidákSD, BH, BY) are given from column 13 to column 20 of the table.
    
    - Significant unadjusted and adjusted p-values are highlighted in yellow and the interpretation of the test is given in the last column of the table.
    
    - Analysis for several or single alleles can be selected through the left panel.
    
    - The table of results is downloadable in a .csv format with possibility of selecting particular columns through the left panel (for the display in the web page the columns 'z', 'rawp', 'Bonferonni', 'Holm', 'Hochberg', 'ŠidákSS', 'ŠidákSD', 'BH' and 'BY' must be selected together in the left panel).
  
  
  • The fourth table display the list of alleles with null or small occurrences. Allele name, Allele type, Nb of IMGT clonotypes (AA), Proportion and Normalized proportion for set 1 and set 2 are given in this table.
    
    For this list of alleles the Fisher's exact test is used and associated p-values are displayed in "rawp" column of the "Test results for alleles" table.
    In this case of null or small allele occurrences
    n_i p_i^k<5 , n_i (1-p_i^k)<5 and n_j p_j^k <5 , n_j (1-p_j^k)<5
    
    where {i,j}_(i≠j): set indexes, n_i (resp., n_j): number of IMGT clonotypes (AA) in set i (resp., set j), k: IMGT allele name and p_i^k (resp., p_j^k): proportion of the allele k in set i (resp. set j), the z-score is not applicable and results displayed in "z" column are not considered.
    
    -Multiple testing procedures are applied for p-values returned by the Fisher's exact test.
    
    -Null or small occurrences figured in this table can be hidden through the left panel.
    
    

4. Multiple testing procedures plots

• The tab 'Multiple testing procedures plots' displays multiple testing procedures visualization plots.
  
  

  4.1. Multiple testing procedures plots for genes

• For the scatter plot (right panel of the figure), negative decimal logarithms (-log₁₀) of unadjusted p-values (black symbols) and adjusted p-values obtained by each multiple testing procedure (colored symbols) and z-scores are reported in a table below the figures.
• The line graphs allow (by hovering with the mouse) the visualization of the exact number of rejected null hypotheses (number of significant differences in proportions) for a chosen Type I error site under a given procedure.
• The scatter plots show negative decimal logarithms (-log10) of unadjusted p-values (black symbols) and adjusted p-values obtained by each multiple testing procedure (colored symbols) against test statistics (z-scores).
• Hovering with the mouse over the scatter plot points allows to see the gene name with the coordinates [procedure (x: z-score, y: -log(p-valeurs,10))].
• If you have no interest to visualize null or small occurrences of genes they can be hidden through the left panel.
  
  
  

  4.2. Multiple testing procedures plots for alleles

  The second figure and table display the same information for multiple testing procedures plots for alleles.
  
  Reminder: Only alleles of genes having significant differences in proportions validated by all multiple testing procedures are analyzed and visualized.
  
  

5. Synthesis graph

• The tab 'Synthesis graph' displays a synthesis graph that combines a normalized bar graph of gene or alleles proportions and the differences in proportions with significance and confidence intervals (CI).
  

  5.1. Synthesis graph for genes

• IMGT gene names visualized in the synthesis graph (y-axis) are ordered by their positions in the locus with the functionality defined for each allele of a gene according to the rules described in the IMGT Scientific chart http://www.imgt.org/IMGTScientificChart/SequenceDescription/IMGTfunctionality.html [6,7,8].
  
  Note that:
  
  
  • Not ordered genes are at the bottom of the gene list (y-axis) with the functionality defined by their alleles and listed in the table below the graph.
    
  • Locus type can be added to the y-axis label through the text input field 'Add locus type to the graph axis title: IGH, IGK, IGL, TRA, TRB, TRG, or TRD' in the left panel (by default the y-axis label is 'genes').
    
    
• Synthesis graph is downloadable in different formats (PNG, JPG and PDF). The height and the width of the graph can be adjusted to improve the clarity of the axes text.
  
  
  

  5.2. Synthesis graph for alleles

  The second figure represents the synthesis graph for alleles.
  
  - No allele order is fixed for this synthesis graph.
  - When several alleles are analyzed (i.e., 'Several genes' option checked through the left panel) the allele names viewed in the synthesis graph are written in a grouped format (e.g., TRBV20-1{*01F,*02F} instead of Homsap TRBV20-1*01 F, or Homsap TRBV20-1*02 F) in order to improve the clarity of the y-axis text.
  The height and the width of the graph can be adjusted for the same reason.
  
  Reminder: Only alleles of genes having significant differences in proportions validated by all multiple testing procedures are analyzed and visualized.

6. CDR-IMGT lengths

7. CDR-IMGT AA properties

• The first table displays by default the frequencies of the 20 amino acids (in rows) at CDR3-IMGT positions (in columns) with the smallest length found in set 1.
  
• CDR-IMGT and the length wanted to be displayed for set 1 can be selected through the left panel.
• Results can be shown in absolute or percentage values selected through the left panel.
• In order to visualize IMGT classes frequencies at a specific CDR-IMGT position, interactive stacked bar graphs were generated for each set:
  - IMGT classes of the 20 common amino acids are showed under the mouse cursor with corresponding frequencies shown between brackets in absolute or percentage values.
  - Possibility to zoom in by clicking and dragging over the graph area if you are interested by a specific range of CDR-IMGT positions (double clicking again to zoom out).
  - Stacked bar graphs are downloadable in PNG format.
• In the figure legend the "X" represents an unidentified amino acid and the "*" represents a stop codon at a specific CDR-IMGT position.
• The height and the width of the graph can be adjusted to improve the clarity of the axes text.

8. V-D-J gene associations

• If the option 'Results with clustering' is checked through the left panel, a double Ward hierarchical clustering with Euclidean distance is performed. This type of classification operates simultaneously on the lines and columns of a matrix intersecting two different types of genes.
  The rows and columns of the central card of the heatmap are ordered such that similar rows and similar columns are close to each other. On vertical and horizontal margins are represented dendrograms that gave rise to these classifications.
• Heatmaps are interactive:
  - For each set, gene label in row is showed under the mouse cursor with its occurrence value with the gene in column.
     If the option "Normalized values" is checked through the left panel, values showed under the mouse cursor are normalized for 10000 IMGT clonotypes (AA) (for IMGT clonotype diversity) or nb of sequences assigned to IMGT clonotypes (AA) (for IMGT clonotype expression).
  - Possibility to zoom in by clicking and dragging over the central card of the heatmap (clicking again to zoom out).
  - Highlight genes in row and/or column by clicking on axis labels.
• If the option 'Results with clustering' is unchecked, heatmaps are shown without dendrograms and ordering.
• Darker colors in heatmaps indicate strong associations between two different IMGT gene types (i.e., higher nb of IMGT clonotypes (AA) (for IMGT clonotype diversity) or higher nb of sequences assigned to IMGT clonotypes (AA) (for IMGT clonotype expression) having the same gene association).
  In contrast, lighter colors indicate genes weakly associated.
• Five color palettes are proposed in the left panel to change the central card color of heatmaps.
• Heatmaps are downloadable in different formats (PNG, JPG and PDF).

E. Exiting IMGT/StatClonotype

The IMGT/StatClonotype tool was developed by Safa Aouinti (IMGT, LIGM, Montpellier, France) using R software (version 3.3.0) and Shiny framework [11].
Interactive graphs are generated using ggplot2 [13] with plotly [14] and d3heatmap [15] packages.

This work was granted access to the HPC resources of CINES under the allocation 2014-036029 made by GENCI (Grand Equipement National de Calcul Intensif) and to Très Grand Centre de Calcul (TGCC) of the Commissariat à l'Energie Atomique et aux Energies Alternatives (CEA) under the allocation 036029 (2010-2016) made by GENCI (Grand Equipement National de Calcul Intensif).

Version 1.0.4 (25/01/2021)

• Fix some result table display issues.
• Update of the gene list and positions in the locus.

Version 1.0.3 (10/02/2018)

• Fix some result table display issues related to the package 'shiny' upgrade (version 1.0.5).
• Some changes in the interface that gives more flexibility to modify the height of the synthesis graphs and improve the clarity of the axes text.
• Add of waiting icons while the script is running.

Version 1.0.2 (03/03/2017)

• Fix some result table display issues related to empty columns in uploaded files.
• Fix some graph display issues related to the package 'plotly' upgrade (version 4.5.6).
• A new dependency package 'colourpicker' is added to avoid warning messages in the R console related to the upgrade of the package 'shinyjs' (version 0.9).
• Close the web browser window and stop the R software process together when you press the button 'Exit IMGT/StatClonotype' (tested with Chrome (version 56.0.2924.87) and Internet Explorer (version 11) web browsers).
• Progressive bars are added while the script is running.

Version 1.0.1 (03/08/2016)

The tab 'Multiple testing procedures plots' now displays interactive scatter plots and interactive line graphs with:

• Genes displayed in scatter plots
• Exact numbers of rejected hypotheses with each multiple testing procedure displayed in line graphs.

• 'IMGTStatClonotype' can be easy updated by typing into the R console each line of code (in red and comments in green) given below:
  
  
  remove.packages("IMGTStatClonotype") # To avoid any conflict between two versions previously installed, please retype this line until this message is displayed in the R console "Error in remove.packages: there is no package called 'IMGTStatClonotype'"
  wspace <- getwd() # Press ENTER
  des =paste(wspace,"/IMGTStatClonotype_1.0.4.tar.gz",sep="") # Press ENTER
  download.file("http://www.imgt.org/download/StatClonotype/IMGTStatClonotype_1.0.4.tar.gz",destfile=des) # Press ENTER
  install.packages("IMGTStatClonotype_1.0.4.tar.gz", repos=NULL) # Press ENTER
  install.packages("colourpicker") # To install the new dependency R package "colourpicker"
  
  
  
• The new version (1.0.4) is now installed. As indicated above, to launch the tool 'IMGT/StatClonotype' you have first to load the package 'IMGTStatClonotype' by typing into R console:
  
  
  library(IMGTStatClonotype) # Press ENTER
  launch() # Press ENTER
  

Created:

Wednesday, 12 August 2015

Last updated:

Monday, 25 January 2021

Authors:

Safa Aouinti safa.aouinti@yahoo.fr, Véronique Giudicelli Veronique.Giudicelli@igh.cnrs.fr, Patrice Duroux Patrice.Duroux@igh.cnrs.fr, Sofia Kossida Sofia.Kossida@igh.cnrs.fr and Marie-Paule Lefranc Marie-Paule.Lefranc@igh.cnrs.fr

Contacts:

Safa Aouinti safa.aouinti@yahoo.fr, Sofia Kossida Sofia.Kossida@igh.cnrs.fr and Marie-Paule Lefranc Marie-Paule.Lefranc@igh.cnrs.fr

IMGT Home page | IMGT Repertoire (IG and TR) | IMGT Repertoire (MH) | IMGT Repertoire (RPI) | IMGT Index | IMGT Scientific chart | IMGT Education

Software material and data coming from IMGT server may be used for academic research only, provided that it is referred to IMGT®, and cited as "IMGT®, the international ImMunoGeneTics information system® http://www.imgt.org (founder and director: Marie-Paule Lefranc, Montpellier, France)." References to cite: Lefranc, M.-P. et al., Nucleic Acids Res., 27:209-212 (1999); doi: 10.1093/nar/27.1.209 Full text Cover; Ruiz, M. et al., Nucleic Acids Res., 28:219-221 (2000); doi: 10.1093/nar/28.1.219 Full text; Lefranc, M.-P., Nucleic Acids Res., 29:207-209 (2001); doi: 10.1093/nar/29.1.207 Full text; Lefranc, M.-P., Nucleic Acids Res., 31:307-310 (2003); doi: 10.1093/nar/gkg085 Full text; Lefranc, M.-P. et al., In Silico Biol., 5, 0006 (2004) [Epub], 5:45-60 (2005); Lefranc, M.-P. et al., Nucleic Acids Res., 33:D593-597 (2005); doi: 10.1093/nar/gki065 Full text; Lefranc, M.-P. et al., Nucleic Acids Res., 37:D1006-1012 (2009); doi: 10.1093/nar/gkn838 Full text; Lefranc, M.-P. et al., Nucleic Acids Res., 43:D413-422 (2015); doi: 10.1093/nar/gku1056 Full text.
For any other use please contact Marie-Paule Lefranc Marie-Paule.Lefranc@igh.cnrs.fr.

     

IMGT® Founder and Executive Director Emeritus:

Marie-Paule Lefranc Marie-Paule.Lefranc@igh.cnrs.fr

IMGT® Director:

Sofia Kossida Sofia.Kossida@igh.cnrs.fr

Bioinformatics manager:

Véronique Giudicelli Veronique.Giudicelli@igh.cnrs.fr

Computer manager:

Patrice Duroux Patrice.Duroux@igh.cnrs.fr

© Copyright 1995-2024 IMGT®, the international ImMunoGeneTics information system® | Terms of use | About us | Contact us