PART I - Processing raw data files

1. NCBI GEO for downloading reads, profiles and peaks from published ChIP-seq experiments

[ETC: 10 mins]

• Open the following publication [PMID:41085408] 

• Go to Related information -> Click on GEO DataSets

• Open the GSE Accession Number [GSE211743] : identify GPL and GSM entries

• Click on the A673,EWSR1-FLI1,R1 (GSM6500452) record

• Identify the Supplementary file section for the genome-wide profile (and/or peaks)

• Search for the SRA link SRX17165694 of the FASTQ raw data

• Press over the SRR21154320 link to access the archive

2. GALAXY for mapping, peak calling, and exporting genome browser tracks

[ETC: 20 mins]

(in X, options to perform whole-genome analysis: this will consume a bigger amount of time)

• Open the Galaxy web platform at this link 

• Register a new account for you and Login

• Open the Get Data section, click on Download and Extract Reads in FASTQ format from NCBI SRA to download the original raw data file SRR21154320

• Click on Open the Upload file from your computer: provide the following FASTQ files (format fastqsanger.gz): EWSR1-FLI1 in A673 [small.fastq.gz] and INPUT in A673 [small2.fastq.gz] 

• Go to Genomic File Manipulation / FASTQ Quality Control and run the FastQC tool on our FASTQ file

• Go to Genomics Analysis and run Bowtie2 on our FASTQ file to perform the mapping on hg38 (use the options Save the bowtie2 mapping statistics to the history and Email notification)

• For each mapping, click over the 👁️ in the Alignments results and inside, on Visualize | display at UCSC (main) to see the reads mapped on the genome. Go to chr7:116465000-116565000

• Go to Genomics Toolkits / deepTools and run the bamCoverage function on each ChIP-seq mapping to generate the genome-wide pile-up of the aligned reads

• For each result, click again over the 👁️ in the bamCoverage output (in UCSC, set the track visualization mode from Dense to Full)

• Look for the MACS2 call peak function in the ChIP-seq subsection of Genomics Analysis. Identify peaks of EWSR1-FLI1 using the INPUT as a control experiment

• Get the published set of peaks of our sample EWSR1-FLI1 in A673 in GEO [GSE211743]. Search for the file GSE211743_DiffBind_FLI1_A673_DPeaks.bed.gz in the Supplementary file table

• Upload the file into GALAXY (bed format, hg19 genome assembly)

• Use the convert formats | CrossMap BED to translate the coordinates of the peaks into hg38

• Use the ✏️ and update the genome database/build of the CrossMap output from hg19 to hg38

• For the visualization, click again over the 👁️ and select display at UCSC (main)

3. UCSC genome browser for visualization

[ETC: 15 mins]

• Click on the 🏠 icon of the UCSC Genome Browser and create a new account for you (My Data/My Sessions or Login)

• Go to the Genome browser screen in hg38

• Switch on the NCBI RefSeq track of genes (Genes and Gene predictions): select the UCSC annotations of RefSeq 

• You should see the following tracks from GALAXY: you can hide the rest of tracks loaded by default

• Click with the Right button of your mouse over each profile and change the track colors

• Configure each bamCoverage track to the maximum height of both to make them comparable

• Generate a PDF screenshot of the region with the custom tracks

• Zoom in and zoom out and move along the chromosome 

• Use the Table browser (in Tools) to calculate the correlation of both profiles in this region

• Finally, open this UCSC session in which we have loaded the genome-wide profiles of EWSR1-FLI1, KDM6A, KDM6B, INPUT (from [PMID:41085408]) and H3K27ac, H3K27me3 and H3K4me3 (from [PMID:25453903])

• Go to the locus of the genes included in the screenshots of the original publication

• Investigate examples of transcriptionally active regions such as CAV1, CDK14, etc.

• Investigate examples of transcriptionally silent regions such as HOXA1, GAD1, etc.

🚀 [TABLE OF CONTENTS OF THE COURSE] 🚀