Annotations
Methylator generates many figures to describe Methylation levels among the differents genomic features. Two files a required:
- A GTF corresponding to genome version used for the mapping
- A texte file with CpG island coordinates
Warning
The genome version specified in the configuration file, in the GTF file, and in the file for CpG island coordinates must be identical.
Tip
Users can also give their own annotation file with coordinates of ChIP-Seq or ATAC-Seq peaks or differentially expressed genes in order to perform multiomics integration analysis
Scheme of the workflow section responsible for building annotations
The R script Annotatr.R is responsible for building the annotations.
To construct these annotations, Methylator relies on several packages:
- annotar
- GenomicRanges
- GenomicFeatures
We have chosen to reconstruct the annotations rather than using annotation databases to avoid dependency on them and to ensure that the annotations are consistently built.
Five gene features are represented:
- Genes
- Intergenic
- Introns
- Exons
- Promoters
Schematic of knownGene annotations (source)
And three types of regions around CGI are represented:
- CpG Islands
- CpG Shores
- CpG Shelves
Schematic of CpG annotations (source)
Annotations description
Genes
Use Genomicfeatures genes function
Intergenic
Intersection of all annotations with genomic annotations. Anything not annotated as a gene is automatically annotated as intergenetic.
Exons
Uses genomicFeatures exonsBy function
Introns
Based on the exons built previously. Considers gaps between exons as introns.
Promoters
Uses the gene annotations constructed above.
Considers a promoter to be + or - 2000 base pairs upstream and downstream of the gene start.
TSS
Create the transcription start sites annotation
tss_gr = IRanges::promoters(genic_gr, upstream = 0, downstream = 1)
GenomicRanges::mcols(tss_gr)$type = sprintf('%s_tss', ORG)
Near TSS
Create the near transcription start sites annotation
5000 base pairs upstream of the TSS and 500 base pairs downstream.
near_tss_gr = IRanges::promoters(genic_gr, upstream = 5000, downstream = 500)
GenomicRanges::mcols(near_tss_gr)$type = sprintf('%s_near_tss', ORG)
Cpg islands
Based on a .BED or .txt annotation file (available on UCSC GoldenPath)
CpG Shores
+/- 2kb upstream and downstream of CpG Islands
CpG Shelves
+/- 4kb upstream and downstream of CpG Islands
The advantage of this approach is that it allows users familiar with the R programming language to modify or add annotations according to their needs.
Warning
When annotating CpGs, DMCs, DMTs, or DMRs, whenever one of these elements overlaps with an annotation, it is associated with that annotation. As a result, the same CpG can be annotated as being located in a gene, a CpG island, and an intron simultaneously. This explains why in the annotation count figures, some categories have more elements than others (e.g., genes). That's why it's also interesting to look at the figures in relative count.
Methylation status in absolut count
Methylation status in relative count
Metadata for user annotation
| annotation_tracks | group | name |
|---|---|---|
| hg19.simplerepeat.txt | repeat | tandem_repeat |
| hg19.nestedRepeats.txt | repeat | nested_repeat |
| hg19.atacseq.txt | atac | atac |
This is necessary only if you select the customs annotations feature in the configuration file. In this case, the workflow will produce, in addition to the figures linked to default annotations, the same figures but for user-provided personal annotations. 'Annotation_tracks' corresponds to the names of the files containing the annotations, group to the different annotation categories, and name to the names that will be given to the annotations in the reports.
# ===== Customs Annotations ===== #
CUSTOM_ANNOT: yes # yes or no
METAFILE_ANNOT: configs/metadata_annot.tsv
CUSTOM_ANNOT_PATH: my_bank/
MERGE_WITH_BASICS_ANNOT: yes # yes or no
Warning
The expected file format for creating custom annotations is as follows :
| chr | start | end | metadata |
|---|---|---|---|
| chr1 | 10000 | 10468 | trf 6 77.2 6 95 3 789 33 51 0 15 1.43 TAACCC |
| chr1 | 10627 | 10800 | trf 29 6 29 100 0 346 13 38 47 0 1.43 AGGCGCGCCGCGCCGGCGC |
| chr1 | 10757 | 10997 | trf 76 3.2 76 95 2 434 17 30 45 6 1.73 GGCGCAGGCGCAGAGAGGC |
| chr1 | 11225 | 11447 | trf 117 1.9 121 80 14 273 12 32 33 20 1.9 CGCCCCCTGCTGGCGAC |
With chromosomal coordinates (chromosomes | start | end) along with any potential supplementary columns.
There is a script called search_bank.sh, in the scripts folder, which, when executed, automatically downloads (when possible) all the necessary annotation files for launching the workflow from UCSC Golden Path. It takes as an argument the name of the reference genome of the species in question. For example:
All annotation files are then downloaded into the folder my_bank provided for this purpose.