How gauravj49/gjsrmap works
Main Components
• Input Data & Reference Files:
– Represent the “input” directory which houses both sequence annotations (under input/annotation/rna_classes) and the reference genomes (under input/reference_genome/dna).
– Indicate that these files serve as both raw sequencing input (fastq files from the user, as indicated via usage instructions) and the custom reference sequences for mapping.
• Pre-processing Module:
– As described in the README (iteration 0), it includes quality control, adapter trimming, removal of low-quality reads, and reference-index building.
– Highlight that this step uses dependency tools (e.g., cutadapt) and scripts (e.g., the pre-processing portions of 00_gjsrmap_usage.sh).
• Iterative Mapping Modules:
– Iteration 1: Mapping filtered reads (16–33 bp) to microRNAs and piRNAs using bowtie.
– Iteration 2: Mapping reads >32 bp to other small RNAs (rRNA, snRNA, snoRNA, premiRNA, osncRNA, etc.).
– Iteration 3: Mapping of reads that remain unmapped to the species’ reference genome.
– Each iteration can be represented as a separate processing block or service.
• Counting and Summarization Module:
– Use components corresponding to the counting-distribution (e.g., get_counts_from_sam.py, scripts/distribute_counts_usage.sh).
– Include QC/summary report generation using FastQC, MultiQC, and plotting via matplotlib.
• Logging and Job Management:
– Note the use of detailed logging and job submission (for example, references like the test/job_submission.log and job directory handling indicated by the JID parameter).
• Helper Scripts / Indexing Tools:
– Include the indexing scripts (STAR_create_index.sh, bowtie_create_index.sh) as supportive components that prepare the data for mapping.
Relationships and Data Flow
• Finally, include a ‘Reporting’ unit that represents generated QC plots (FastQC, MultiQC, matplotlib-based summary plots) and logs.
Architectural Patterns and Principles
• Modularity: Emphasize that each mapping iteration and pre-processing step are separate modules/scripts, which promotes a modular design.
• Script-based Workflow Orchestration: Note that a wrapper script (e.g., 06_run_ncRNA_mapping_usage.sh) ties these modules together, so it should be represented as an orchestrator that initiates the pipeline with user-specified parameters.
Relevant Technologies and Dependencies
• The usage of Python scripts (ncRNA_mapping_summary.py, get_counts_from_sam.py, etc.) should be shown in the Processing or Summarization modules.