QuagmiR configuration - Gu-Lab-RBL-NCI/QuagmiR GitHub Wiki

Configuration of QuagmiR runs is specified under the config.yaml file.

FILTERING

Reads pulled based on the miRNA_motif are further filtered to be considered under >miRNA_name based on their sequence similarity to miRNA_consensus_sequence. This comparison take place on the nucleotide that precede and follow the miRNA_motif.

It is well known that isomiRs present an asymmetrical distribution of the sequence heterogeneity, due to trimming and tailoring events occuring disproportionately on the 3' end. In order to capture this biological diversity while also preserving an accurate mapping of the reads, an independent 5' and 3' filtering parameter is included. The sequence similarity is calculated using the Levenshtein distance obtained from (https://github.com/infoscout/weighted-levenshtein). Distances used for insertions, deletions, and transversions, can be further customized if necessary under # DISTANCE METRIC, although by default any change is set to 1.

edit_distance_3p: 3 # to be set to -1 to switch off 3p filtering
edit_distance_5p: 2 # to be set to -1 to switch off 5p filtering

We can define with min_ratio the minimal percentage that a given isomiR requires to be considered in the sequence report isomir.sequence_info.tsv.

min_ratio: .1 # to be set to -1 to switch off min_ratio filtering

Similarly, we can define with min_read the minimal number of counts that a given isomiR requires to be considered in the sequence report isomir.sequence_info.tsv.

min_read: 9 # to be set to -1 to switch off min_read filtering

DISPLAY

This section allow the user to decide whether or not to generate various output reports:

Summary of global outputs isomir.tsv

display_summary: True

Detailed sequence of the isomiR sequences sequence_info.tsv

display_sequence_info: True

Detailed report on nucleotide composition by position nucleotide_dist.tsv

display_nucleotide_dist: True

Note that this report only provides information for nucleotides beyond the miRNA_motif

Combined summary of global outputs isomir.tsv found under group_results**

display_group_output: True

Display of counts per million (CPM) and reads per kilobase per million reads (RPKM) on the reports

display_expression: True

Note that the use of CPM is generally preferred given the relative similarity in size of mature miRNA.

Display distance metric on the reports

display_distance_metric: True

AMBIGUOUS LETTERS

In order to increase the sensitivity to RNA editing events that may happen in the miRNA, the miRNA_motif allow the usage of ambiguous nucleotides by ambiguous_letters: True. Further details of this feature can be found under the motif section. Note that running QuagmiR on this mode increases its runtime. The activation of this mode also allows the matching of eventual sequence with ambiguous nucleotides from the fastq file.

DESTRUCTIVE MOTIF PULL

The function destructive_motif_pull when True allows to assign reads that match for more than one motif to the one that presents the least distance to the full miRNA sequence.

INPUT

The default pathway to access the input files can be edited using the following field:

data: 'data' # path to folder containing input files

The motif_consensus_file: establishes the input references that will be taken by QuagmiR to perform the sequence motif pull and posterior isomiR analysis.

motif_consensus_file: 'motif-consensus.fa'

The reference_file: is required by QuagmiR to establish the hairpin and coordinates needed in the generation of the GFF report. mirbase_version: and source_ontology: are onely required to fill the text field displayed at the header of the GFF report.

reference_file: 'miRBase21-master.tsv'
mirbase_version: '21' # version of miRBase being used for reference files
source_ontology: 'miRBase v21 doi:10.25504/fairsharing.hmgte8'

OUTPUT

The group_output_name: allows the user to aggregate results from multiple samples into a single report file. By default the name of this aggregated file is set to cohort1.

CUSTOMIZED DISTANCE METRICS

The distance matrix allow the user to change the penalty or distance score assigned to each possible variant on a read compared to the reference sequence.

deletion_score: 1
insertion_score: 1
substitution_AG: 1 # transition
substitution_GA: 1 # transition
substitution_CT: 1 # transition
substitution_TC: 1 # transition
substitution_AT: 1 # transversion
substitution_TA: 1 # transversion
substitution_AC: 1 # transversion
substitution_CA: 1 # transversion
substitution_GC: 1 # transversion
substitution_CG: 1 # transversion
substitution_GT: 1 # transversion
substitution_TG: 1 # transversion