Explore Workflows

This ATAC pipeline is based on original [BioWardrobe's](https://biowardrobe.com) [PubMed ID:26248465](https://www.ncbi.nlm.nih.gov/pubmed/26248465) **ChIP-Seq** basic analysis workflow for a **paired-end** experiment with Trim Galore. The pipeline was adapted for ATAC-Seq single-read data analysis by updating genome coverage step. ### Data Analysis Steps SciDAP starts from the .fastq files which most DNA cores and commercial NGS companies return. Starting from raw data allows us to ensure that all experiments have been processed in the same way and simplifies the deposition of data to GEO upon publication. The data can be uploaded from users computer, downloaded directly from an ftp server of the core facility by providing a URL or from GEO by providing SRA accession number. Our current pipelines include the following steps: 1. Trimming the adapters with TrimGalore. This step is particularly important when the reads are long and the fragments are short as in ATAC -resulting in sequencing adapters at the end of read. If adapter is not removed the read will not map. TrimGalore can recognize standard adapters, such as Nexterra/Tn5 adapters. 2. QC 3. (Optional) trimming adapters on 5' or 3' end by the specified number of bases. 4. Mapping reads with BowTie. Only uniquely mapped reads with less than 3 mismatches are used in the downstream analysis. Results are saved as a .bam file. 5. Reads mapping to chromosome M are removed. Since there are many copies of chromosome M in the cell and it is not protected by histones, some ATAC libraries have up to 50% of reads mapping to chrM. We recommend using OMNI-ATAC protocol that reduces chrM reads and provides better specificity. 6. (Optional) Removal of duplicates (reads/pairs of reads mapping to exactly same location). This step is used to remove reads overamplified in PCR. Unfortunately, it may also remove \"good\" reads. We usually do not remove duplicates unless the library is heavily duplicated. Please note that MACS2 will remove 'excessive' duplicates during peak calling ina smart way (those not supported by other nearby reads). 7. Peakcalling by MACS2. (Optionally), it is possible to specify read extension length for MACS2 to use if the length determined automatically is wrong. 8. Generation of BigWig coverage files for display on the browser. Since the cuts by the Tn5 transposome are 9bp apart, we show coverage by 9bp reads rather than fragments as in ChIP-Seq. The coverage shows the number of fragments at each base in the genome normalized to the number of millions of mapped reads. This way the peak of coverage will be located at the most accessible site. ### Details _Trim Galore_ is a wrapper around [Cutadapt](https://github.com/marcelm/cutadapt) and [FastQC](http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) to consistently apply adapter and quality trimming to FastQ files, with extra functionality for RRBS data. A [FASTQ](http://maq.sourceforge.net/fastq.shtml) input file has to be provided. In outputs it returns coordinate sorted BAM file alongside with index BAI file, quality statistics for both the input FASTQ files, reads coverage in a form of BigWig file, peaks calling data in a form of narrowPeak or broadPeak files, islands with the assigned nearest genes and region type, data for average tag density plot (on the base of BAM file). Workflow starts with running fastx_quality_stats (steps fastx_quality_stats_upstream and fastx_quality_stats_downstream) from FASTX-Toolkit to calculate quality statistics for both upstream and downstream input FASTQ files. At the same time Bowtie is used to align reads from input FASTQ files to reference genome (Step bowtie_aligner). The output of this step is unsorted SAM file which is being sorted and indexed by samtools sort and samtools index (Step samtools_sort_index). Depending on workflow’s input parameters indexed and sorted BAM file could be processed by samtools rmdup (Step samtools_rmdup) to remove all possible read duplicates. In a case when removing duplicates is not necessary the step returns original input BAM and BAI files without any processing. If the duplicates were removed the following step (Step samtools_sort_index_after_rmdup) reruns samtools sort and samtools index with BAM and BAI files, if not - the step returns original unchanged input files. Right after that macs2 callpeak performs peak calling (Step macs2_callpeak). On the base of returned outputs the next step (Step macs2_island_count) calculates the number of islands and estimated fragment size. If the last one is less that 80 (hardcoded in a workflow) macs2 callpeak is rerun again with forced fixed fragment size value (Step macs2_callpeak_forced). If at the very beginning it was set in workflow input parameters to force run peak calling with fixed fragment size, this step is skipped and the original peak calling results are saved. In the next step workflow again calculates the number of islands and estimated fragment size (Step macs2_island_count_forced) for the data obtained from macs2_callpeak_forced step. If the last one was skipped the results from macs2_island_count_forced step are equal to the ones obtained from macs2_island_count step. Next step (Step macs2_stat) is used to define which of the islands and estimated fragment size should be used in workflow output: either from macs2_island_count step or from macs2_island_count_forced step. If input trigger of this step is set to True it means that macs2_callpeak_forced step was run and it returned different from macs2_callpeak step results, so macs2_stat step should return [fragments_new, fragments_old, islands_new], if trigger is False the step returns [fragments_old, fragments_old, islands_old], where sufix \"old\" defines results obtained from macs2_island_count step and sufix \"new\" - from macs2_island_count_forced step. The following two steps (Step bamtools_stats and bam_to_bigwig) are used to calculate coverage on the base of input BAM file and save it in BigWig format. For that purpose bamtools stats returns the number of mapped reads number which is then used as scaling factor by bedtools genomecov when it performs coverage calculation and saves it in BED format. The last one is then being sorted and converted to BigWig format by bedGraphToBigWig tool from UCSC utilities. To adapt the pipeline for ATAC-Seq data analysis we calculate genome coverage using only the first 9 bp from every read. Step get_stat is used to return a text file with statistics in a form of [TOTAL, ALIGNED, SUPRESSED, USED] reads count. Step island_intersect assigns genes and regions to the islands obtained from macs2_callpeak_forced. Step average_tag_density is used to calculate data for average tag density plot on the base of BAM file.

https://github.com/datirium/workflows.git

Path: workflows/trim-atacseq-pe.cwl

Branch/Commit ID: f3e44d3b0f198cf5245c49011124dc3b6c2b06fd

https://github.com/genome/analysis-workflows.git

Path: definitions/subworkflows/sv_paired_read_caller_filter.cwl

Branch/Commit ID: 2f65fc96207a71b1cda4e246f808bed056608cd0

Sequence reads are first cleaned from adapters and transformed into fully bisulfite-converted forward (C->T) and reverse read (G->A conversion of the forward strand) versions, before they are aligned to similarly converted versions of the genome (also C->T and G->A converted). Sequence reads that produce a unique best alignment from the four alignment processes against the bisulfite genomes (which are running in parallel) are then compared to the normal genomic sequence and the methylation state of all cytosine positions in the read is inferred. A read is considered to align uniquely if an alignment has a unique best alignment score (as reported by the AS:i field). If a read produces several alignments with the same number of mismatches or with the same alignment score (AS:i field), a read (or a read-pair) is discarded altogether. On the next step we extract the methylation call for every single C analysed. The position of every single C will be written out to a new output file, depending on its context (CpG, CHG or CHH), whereby methylated Cs will be labelled as forward reads (+), non-methylated Cs as reverse reads (-). The output of the methylation extractor is then transformed into a bedGraph and coverage file. The bedGraph counts output is then used to generate a genome-wide cytosine report which reports the number on every single CpG (optionally every single cytosine) in the genome, irrespective of whether it was covered by any reads or not. As this type of report is informative for cytosines on both strands the output may be fairly large (~46mn CpG positions or >1.2bn total cytosine positions in the human genome).

https://github.com/datirium/workflows.git

Path: workflows/bismark-methylation-se.cwl

Branch/Commit ID: ddc35c559d1ac6aab4972fe1a2b63300c4373f54

https://github.com/ncbi/pgap.git

Path: task_types/tt_kmer_ref_compare_wnode.cwl

Branch/Commit ID: f403d9e26d60d3e3591a03077bc9dfa188b1c2bb

https://github.com/cnherrera/CWL_Workflow_DIADEM_use_case.git

Path: work9.cwl

Branch/Commit ID: 2932d26107eb418c1adf865cc7c13c147a6ed71d

https://github.com/common-workflow-language/cwltool.git

Path: tests/wf/scatter2.cwl

Branch/Commit ID: d5f7fa162611243f0c66dd3e933c16a4964a09ca

https://git.astron.nl/rd/vlbi-cwl.git

Path: workflows/concatenate-flag.cwl

Branch/Commit ID: aea458f090644ebfbee6ba9be992ed308bb583e0

https://github.com/genome/analysis-workflows.git

Path: definitions/subworkflows/qc_exome.cwl

Branch/Commit ID: 0c4f4e59c265eb22aed3d2d37b173cb5430773d2

https://github.com/genome/analysis-workflows.git

Path: definitions/pipelines/alignment_wgs_mouse.cwl

Branch/Commit ID: adcae308fdccaa1190083616118dfadb4df65dca

Workflow runs [Bowtie](http://bowtie-bio.sourceforge.net/tutorial.shtml) v1.2.0 (12/30/2016) to build indices for reference genome provided in a single FASTA file as fasta_file input. Generated indices are saved in a folder with the name that corresponds to the input genome

https://github.com/datirium/workflows.git

Path: workflows/bowtie-index.cwl

Branch/Commit ID: 57863b6131d8262c5ce864adaf8e4038401e71a2