Infinium Human Methylation BeadChip
Which DNA regions and positions are diffrentialy methylated in pre MAPKi treatment and post MAPKi resistance Melanomas GSE65183?
How to analyse and visualise Infinium Human Methylation BeadChip’s?
Learn how to perform reproducible Infinium Human Methylation BeadChip analysis
Visualise differentially methylated positions using UCSC browser
- Introduction to Galaxy Analyses
- Sequence analysisTime estimation: 1 hourSupporting Materials:Published: Aug 27, 2018Last modification: Nov 3, 2023License: Tutorial Content is licensed under Creative Commons Attribution 4.0 International License. The GTN Framework is licensed under MITpurl PURL: https://gxy.io/GTN:T00139rating Rating: 3.0 (0 recent ratings, 2 all time)version Revision: 62
This tutorial is based on Hugo W, Shi H, Sun L, Piva M et al.: Non-genomic and Immune Evolution of Melanoma Acquiring MAPKi Resistance Hugo et al. 2015.
In this tutorial we will do:
- Raw intensity data loading
- .idat preprocessing
- Differentially methylated regions and positions analysis
- Annotation and visualization
We will use a small subset of the original data. If we run the tutorial on the orginal dataset, analysis will be time consuming and not reproducible Infinium Human Methylation BeadChip computation on the orginal data can be found at case study.
The field of cancer genomics has demonstrated the power of massively parallel sequencing techniques to inform on genes and specific alterations that drive tumor onset and progression. Although large comprehensive sequence data sets continue to be made increasingly available, data analysis remains an ongoing challenge, particularly for laboratories lacking dedicated resources and bioinformatics expertise. To address this, we have provided training based on Galaxy Infinium Human Methylation BeadChip tool that represents many popular algorithms for detecting somatic genetic alterations from genome and exome data.
This exercise uses datasets from the Cell publication by Hugo et al. 2015. with the goal being the identification of differentially methylated regions and positions associated with treatment resistant melanomas. Datasets include the Infinium Human Methylation BeadChip array performed in melanoma tumors in a sample of patients pre and post MAPKi and BRAFi treatment with different outcomes (sensitive and resistant). For each sample there is raw green (methylated) and red (unmethylated) colour arrays containing the summarised bead information generated by the Infinium Human Methylation BeadChip scanner.
The Infinium Human Methylation BeadChip uses two different bead types to detect changes in DNA methylation levels. In the figure we can see M - methylated and U - unmethylated bead types. In our study unmethylated and methylated bead signals are reported as green and red colors respectively.
The workflow combines 5 main steps, starting with raw intensity data loading (.idat) and then optional preprocessing and normalisation of the data. The next quality control step performs an additional sample check to remove low-quality data, which normalisation cannot detect. The workflow gives the user the opportunity to perform any of these preparation and data cleaning steps, including a highly recommended genetic variation annotation step resulting in single nucleotide polymorphism identification and removal. Finally, the dataset generated through all of these steps can be used to hunt (find) differentially-methylated positions (DMP) and regions (DMR) with respect to a phenotype covariate.
Raw intensity data loading
The first step of the Infinium Human Methylation BeadChip array analysis is raw methylation data loading (intensity information files for each two colour micro array)
Hands-on: Data Loading
Create a new history for this tutorial and give it a proper name
Click the new-history icon at the top of the history panel.
If the new-history is missing:
- Click on the galaxy-gear icon (History options) on the top of the history panel
- Select the option Create New from the menu
- Import the following IDAT files from Zenodo or from the data library (ask your instructor)
https://zenodo.org/record/1251211/files/GSM1588704_8795207135_R01C02_Red.idat https://zenodo.org/record/1251211/files/GSM1588706_8795207135_R02C02_Red.idat https://zenodo.org/record/1251211/files/GSM1588705_8795207119_R05C02_Red.idat https://zenodo.org/record/1251211/files/GSM1588707_8795207119_R06C02_Red.idat https://zenodo.org/record/1251211/files/GSM1588704_8795207135_R01C02_Grn.idat https://zenodo.org/record/1251211/files/GSM1588706_8795207135_R02C02_Grn.idat https://zenodo.org/record/1251211/files/GSM1588705_8795207119_R05C02_Grn.idat https://zenodo.org/record/1251211/files/GSM1588707_8795207119_R06C02_Grn.idat https://zenodo.org/record/1251211/files/phenotypeTable.txt
- Copy the link location
Click galaxy-upload Upload Data at the top of the tool panel
- Select galaxy-wf-edit Paste/Fetch Data
Paste the link(s) into the text field
- Close the window
As an alternative to uploading the data from a URL or your computer, the files may also have been made available from a shared data library:
- Go into Shared data (top panel) then Data libraries
- Navigate to the correct folder as indicated by your instructor.
- On most Galaxies tutorial data will be provided in a folder named GTN - Material –> Topic Name -> Tutorial Name.
- Select the desired files
- Click on Add to History galaxy-dropdown near the top and select as Datasets from the dropdown menu
In the pop-up window, choose
- “Select history”: the history you want to import the data to (or create a new one)
- Click on Import
Phenotype table can be in different sizes with different arguments, however the second column is required to contain phenotype covariate information for each sample.
- Run UCSC Main tool to obtain the reference genome. The tool will take you to the UCSC table browser. Use the following parameters to extract the reference genome
Feb. 2009 (GRCh37/hg19)
- “output format”:
GTF - gene transfer (limited)
- “Send output to”:
- Click on the get output button at the bottom of the screen
- On the next page, click on the Send Query to Galaxy button
- Wait for the upload to finish
After exporting the reference genome from UCSC, we need to make sure that it is in the right dataset build.
Click on the Differentially_Methylated_Positions.bed output in your history to expand it.
Set the database build of your dataset to
Human Feb. 2009 (GRCh37/hg19) (hg19)(if it is not set automatically)
- Click on the galaxy-pencil pencil icon for the dataset to edit its attributes
- In the central panel, change the Database/Build field
- Select your desired database key from the dropdown list:
- Click the Save button
display at UCSCtowards the bottom of the history item. This will launch UCSC Genome Browser with your Custom Track
Preprocessing and data quality assurance is an important step in Infinium Methylation Assay analysis. Idat dataset represents two colour data with a green and a red channel and can be converted into methylated and unmethylated signals or into Beta values. The Infinium Human Methylation BeadChip tool extracts and plots the quality control data frame with two columns mMed and uMed which are the medians of methylation signals (Meth and Unmeth). Comparing them against one another allows users to detect and remove low-quality samples.
If your files require normalisation, you might prefer to use one of the other preprocessing tools provided in Infinium Human Methylation BeadChip tool i.e. Preprocess Funnorm or Preprocess Quantile look for recommendation at Aryee et al. 2014.
Differentially methylated regions and positions analysis
The main goal of the Infinium Human Methylation BeadChip analysis is to simplify the way differentially methylated loci sites are detected. The Infinium Human Methylation BeadChip pipeline contains differentially methylated positions (DMPs) detection with respect to a phenotype covariate, and more complex solutions for finding differentially methylated regions (DMRs). Genomic regions that are differentially methylated between two conditions can be tracked using a bumphunting algorithm. The algorithm first implements a t-statistic at each methylated loci location, with optional smoothing, then groups probes into clusters with a maximum location gap and a cutoff size to refer the lowest possible value of genomic profile hunted by our tool.
Hands-on: detecting methylated loci sites
- Run Infinium Human Methylation BeadChip ( Galaxy version 2.1.0) with the following parameters to map the imported datasets against phenotype covariate and reference genome obtained from UCSC.
- param-files “red channel files”: all files ending in
- param-files “green channel files”: all files ending in
Ilumina methylation array data can be mapped to the genome with or without additional preprocessing methods. Incomplete annotation of genetic variations such as single nucleotide polymorphism (SNP) may affect DNA measurements and disrupt downstream analysis of results. Aryee et al. 2014 It is highly recommended to remove the probes that contain either an SNP at the methylated loci interrogation or at the single nucleotide extension. In this tutorial we will remove probes affected by genetic variation by selecting (Optional) Preprocessing Method tool.
“(Optional) Preprocessing Method”:
- “Phenotype Table”:
The phenotypeTable.txt file uploaded from Zenodo
- “maxGap Size”:
250We will use the default gap of 250 base pairs (bps), i.e. any two points more than 250 bps away are put in a new cluster.
- “Cutoff Size”:
0.1In order to find segments that are positive, near zero, and negative. We need a cutoff which is one number in which case “near zero” default 0.1
- “Number of Resamples”:
0Default value 0 for permutation method apply selection of randomized cases with replacement from the original data while using ‘bootstrap’ method.
permutationMethod used to generate null candidate regions, must be one of ‘bootstrap’ or ‘permutation’ (defaults to ‘permutation’).
- “Phenotype Type”:
categoricalIdentify regions where methylation is associated with a continuous or categorical phenotype.
- “qCutoff Size”:
0.5Diffrentialy methylated positions with an FDR q-value greater than this value will not be returned.
- “Variance Shrinkage”:
TRUEDefault TRUE as it is recommended when sample sizes are small <10
- “Genome Table”:
How do we define phenotype covariate?
Phenotype covariate is the set of observable characteristics of an individual resulting from the gene-environment interactions
Annotation and visualization
In addition to downstream analysis users can annotate the differentially methylated loci at the promoter regions of genes with gene function descriptions, and relationships between these concepts.
Hands-on: Annotate Differentially Methylated Position
- Run ChIPpeakAnno annoPeaks ( Galaxy version 0.1.0) on the output of Infinium Human Methylation BeadChip ( Galaxy version 2.1.0) with the following parameters
- param-file “Differentialy methylated Positions”: output of Infinium Human Methylation BeadChip ( Galaxy version 2.1.0)
“Additional Column of Score”:
Position of column of score optional value if it is required
- Run Cut on the previous output adjusting the following parameters to cut “gene_name” column from table of annotated peaks and then get a list of genes
- “Cut columns”:
- “Delimited by”:
- param-file “From”: output of ChIPpeakAnno annoPeaks ( Galaxy version 0.1.0)
- Use Remove beginning on
Gene Listwith the following parameters
- “Remove first”:
- param-file “from”: output of Cut
- Run Cluster Profiler Bitr ( Galaxy version 0.1.0) on the previous output adjusting the following parameters to convert the list of genes to list of entrez ID
- “Input Type Gene ID”:
- “Output Type Gene ID”:
- Run a GO Enrichment Analysis using clusterProfiler go ( Galaxy version 0.1.0) on the output of the Cluster Profiler Bitr ( Galaxy version 0.1.0)
|PTGS2 / KCNC1 / FZD1 /SLC22A18 /SLC22A3 (…)
|response to peptide
|SULF1/ LAMA5/ MED1 /CFLAR/ MSX2 (…)
|response to steroid hormone
|HDAC9/ RAB10/ CFLAR/ WDTC1 (…)
Epigenetic aberrations which involve DNA modifications give researchers an interest in identifying novel non-genetic factors responsible for complex human phenotypes such as height, weight, and disease. To identify methylation changes researchers need to perform complicated and time consuming computational analysis. Here, the EWAS suite becomes a solution for this inconvenience and provides a simplified downstream analysis available as a ready to run pipline in supplementary materials.