Use of genes

[cavelandiah]

Dear @y9c,

Recently, I developed a tailored analysis using pseudoU-BIDseq to analyze my custom-aligned BAM files. I found this reference in the documentation describing how to use pre-analyzed BAM files, which was very helpful.

I am currently combining independent mapping sources in my project:

• One alignment with STAR to the full genome
• Another with bowtie2 to custom gene regions.

I configured the config.yaml like this:

reference:
  genes:
    fa: /data/geneRef/mm10_regions_masked_extended.fa
  genome:
    fa: /data/reference/mm10_masked.fa
    star: /data/mm10_masked_STAR

samples:
  AAR-WT_Control_1:
    bam:
      genes: /mapping_regions_default_classified/MJA16_S5_region_sorted.bam
      genome: /mapping_star/MJA16_S5_genome_sorted.bam
    group: AAR-WT_Control
    treated: false
AAR-WT_BS_1:
    bam:
      genes: /mapping_regions_default_classified/MJA17_S6_region_sorted.bam
      genome: /mapping_star/MJA17_S6_genome_sorted.bam
    group: AAR-WT_BS
    treated: true

forward_stranded: false

The pipeline executes successfully and produces two sets of outputs:

├── call_sites
│   ├── genes.tsv.gz
│   └── genome.tsv.gz
└── filter_sites
    ├── genes.tsv.gz
    └── genome.tsv.gz

And final table in filter_sites/genes.tsv.gz looks like this:

chr     pos     strand  AAR-WT_Control_1_depth AAR-WT_Control_1_gap   AAR-WT_BS_1_depth      AAR-WT_BS_1_gap        AAR-WT_Control_ratio   AAR-WT_BS_ratio        AAR-WT_Control_fraction        AAR-WT_BS_fraction     AAR-WT_Control_passed  AAR-WT_BS_passed
Pseudo_site_22838::chr10:117053989-117054190::Rw        101     +       1319    0       298     9       NaN     0.030   NaN     0.094   0       1
Pseudo_site_13135::chr10:127096542-127096743::Rw        21      +       548     0       150     8       NaN     0.053   NaN     0.173   0       1
Pseudo_site_16602::chr10:128485571-128485772::Fw        62      +       132     0       100     11      NaN     0.110   NaN     0.563   0       1

Based on this output, I have a few questions:

• Does pseudoU-BID-seq apply the same analysis logic to the genome and genes references independently? That is, are the detection and filtering steps performed in parallel for each reference?
• Are the genomic coordinates reported in the output 0-based or 1-based? This is important for my downstream annotation.
• What does the *_passed column mean in this context? Does a value of 1 indicate that a site passed a certain quality or statistical threshold?
• I am interested in the intermediate results to generate these tables, is there any option to get them?

Thank you again for this excellent tool for Pseudouridines detection and for making advanced customization possible. I look forward to your response!

[y9c]

Hi @cavelandiah,

Thank you for using this pipeline.

The reply to your questions is as below.

• Does pseudo-BID-seq apply the same analysis logic to the genome and genes references independently? That is, are the detection and filtering steps performed in parallel for each reference?

Yes, the cutoff for gene and genome are the same, and they can run in pralllel.

• Are the genomic coordinates reported in the output 0-based or 1-based? This is important for my downstream annotation.

The coordinates are 1-based.

• What does the *_passed column mean in this context? Does a value of 1 indicate that a site passed a certain quality or statistical threshold?

Yes, the _passed column indicate this site passed all the cutoff, including depth, ratio, and p-value..

• I am interested in the intermediate results to generate these tables, is there any option to get them?

The intermediate results are removed automatically, because they are marked as temp, but you can keep them by add --no-temp argument to save them.