Analyzing PacBio sequel data and SRA

Question

[Smandape]

I am new to NCBI's SRA and pacbio sequencing data analysis. I am trying to download pacbio sequence data from SRA and my goal is get CCS (Consensus) reads [fastq/a format is best]. The data is generated from PacBio's sequel instrument. What I read so far relates to the data analysis methods from their previous instrument (RS II). I referred to online blogs and Biostars posts (such as this and this). My questions are related to both SRA and pacbio sequence data:

1. What do runs mean in SRA (for example HG00733 ) when filtered for (Source: DNA, Platform: PacBioSMRT) on right yields 7 experiments. Some of them have multiple runs for an experiment. For example, this is a WGS on Sequel II and has 7 runs. The only difference I found between these runs is different library name but for other experiments with multiple runs I didn't find anything different between runs.

2. I am particularly interested in this experiment (https://www.ncbi.nlm.nih.gov/sra/SRX4480530[accn] ). I want to get consensus reads for this experiment. What I read from other posts say it is easy to start with raw data or subreads files. I went to the 'Data access' page for this experiment's run but there are multiple subreads.bam files in the 'Original format' (https://trace.ncbi.nlm.nih.gov/Traces/sra/?run=SRR7615963 ). My question is what does it mean when there are multiple subreads.bam files? And how do I get consensus reads from multiple bam files? I read about SMRTlink binary software and Canu to process subreads.bam to get consensus. Which tool should I use here? Or is there a direct way to get consensus reads from SRA?

3. For the experiment mentioned above (SRX4480530), the original format bam files say subreads.bam but for other experiments (such as https://trace.ncbi.nlm.nih.gov/Traces/sra/?run=ERR3822935 ) the bam file doesn't say subreads in its filename. In such case what file is it?

[GenoMax]

1. When filtered for PacBio SMRT (in left hand column) I see 10 experiments that includes HG00733 samples. There is one Sequel and two Sequel II experiments. Remaining are older RS runs.

2. I think PacBio put a certain number of reads in each file in order to avoid creating a gigantic BAM file. You can easily check that. There is no direct way to get consensus reads from these AFAIK. You will need to use programs you mentioned or SequelTools (LINK) which I just came across.

3. PacBio uses the following BAM name conventions for different types of files

---

Data type                           Filename template
ZMW reads from movie                movieName.zmws.bam

Analysis-ready subreads 1           movieName.subreads.bam
    from movie

Excised adapters, barcodes, and     movieName.scraps.bam
    rejected subreads

CCS reads computed from movie       movieName.ccs.bam
Aligned subreads in a job           jobID.aligned_subreads.bam
Aligned CCS in a job                jobID.aligned_ccs.bam

[Smandape]

@genomax Thank you for reply. I went through the BAM name conventions but this Sequel II experiment's ( https://trace.ncbi.nlm.nih.gov/Traces/sra/?run=ERR3822935 ) bam filename doesn't match any of the mentioned templates. What that would mean in such case? Also, do different runs in an experiment mean replicates or maybe under different conditions those were sequenced for that experiment? Thanks!