BGCflow v1 scope

[matinnuhamunada]

@OmkarSaMo I think we should define the scope of the workflow (at least for the first version/release). This will help us to focus on delivering a stable release.

I think this tagline you wrote is cool: "Snakemake workflow to systematically analyze BGCs and pangenomes of large number genomes" Which then let us define the rules required for the first stable release.

As more tools are being added, it is hard to keep track of things that are happening in the workflow. Categorizing the rules and giving a simple definition of what the rules should do will help us focus and find out what tools are still missing or prioritized.

Public database utilities:

• [x] ncbi-download.smk → fetch fasta from ncbi for given RefSeq assembly accession list
• [x] patric.smk → fetch fasta from patrick for given PATRIC accession list

Pre-processing & QC:

• [x] seqfu.smk → assembly statistics and QC

Phylogenomic analysis:

• [x] gtdb.smk → taxonomic placement using publicly available GTDB taxonomy through API
• [x] refseq_masher.smk → fetch top 10 closest relative from public NCBI RefSeq database
• [x] mlst.smk → PubMLST typing
• [x] automlst_wrapper.smk → phylogenetic tree build from set of genomes in given project

Gene annotation and Enrichment:

• [x] prokka.smk → gene calling and annotation
• [x] eggnog.smk → functional enrichment

BGC analysis:

• [x] antismash.smk → mine genome for BGCs
• [x] bigscape.smk → compare BGCs across set of genomes in given project

Pangenome analysis:

• [x] roary.smk → build pangenome of genomes in given project

Summary & reporting:

• [x] bgc_analytics.smk → summarize raw data to integrate with notebooks for EDA & data mining

I made a tick to the minimal functions that is required to run the workflow. What do you think?

PS: this could be nice for the docs / wiki

[OmkarSaMo]

How about following?

Scope:

"Large-scale automated and adaptive workflows to systematically analyze BGCs and pangenomes of various bacterial groups"

Vision:

"Our vision is to computationally characterize and access any cluster, from every genome"

[OmkarSaMo]

I have edited some of the rule categories and descriptions. Will check out again later