Comprehensive summary: GEO vs NCBI Submission Portal (SRA/BioProject/BioSample) for bulk RNA-seq + GeoMx DSP

https://submit.ncbi.nlm.nih.gov/subs/

https://www.ncbi.nlm.nih.gov/geo/subs/

1) What each platform is best for

  • NCBI Submission Portal Used primarily to submit raw sequencing reads to the Sequence Read Archive (SRA) and to create/link the organizing records:

    • BioProject (the overall study / project container)
    • BioSample (the per-sample metadata records required for SRA)
  • NCBI GEO (Gene Expression Omnibus) Best for gene expression–style datasets and processed outputs, including:

    • count matrices / processed tables
    • sample/ROI metadata tables
    • GeoMx DSP outputs (DCC/PKC)
    • supplementary analysis outputs and documentation (README, workflow description) GEO is commonly used for “processed + metadata”, while raw FASTQs (if available) go to SRA.

2) Where to submit your two dataset types

  • Bulk RNA-seq (Illumina) with raw FASTQ files

    • Submit to SRA via the NCBI Submission Portal.
    • Ensure you have/define: BioProject + BioSamples, then upload FASTQs and metadata.
  • GeoMx DSP spatial transcriptomics with only DCC/PKC + annotation Excel (+ R workflow), no FASTQs

    • Submit to GEO (not SRA), because DCC/PKC are processed outputs, not raw reads.
    • In GEO, upload:

      • DCC files (processed counts + QC/metrics per ROI/AOI)
      • PKC files (panel/probe definitions)
      • annotation table (Excel is okay; CSV/TSV is even better for machine readability)
      • README describing file structure + column meanings + mapping between ROI IDs and metadata
      • analysis workflow (e.g., R scripts; optionally hosted on GitHub/Zenodo and linked)
  • If GeoMx FASTQs are obtained later

    • Submit GeoMx FASTQs to SRA, and in the GEO record state: raw reads in SRA, processed data in GEO, cross-linking accessions.

3) What to choose in the NCBI “Start a new submission” list

For your use case:

  • Choose Sequence Read Archive (SRA) ✅ for FASTQ (raw reads).
  • You will also use/associate BioProject and BioSample (often created during the SRA workflow).
  • Do not use GenBank/TSA/Genome for these transcriptomics read submissions (those are for assembled sequences/genomes, not raw RNA-seq reads).
  • GEO is not started from that list; it has its own submission entry point.

4) Accounts / login clarification

  • Same login principle: In general, you use the same My NCBI account across NCBI services.
  • Why GEO can feel like “another account”:

    • GEO requires a GEO Submitter Profile (contact identity/profile) attached to a My NCBI account.
    • In many labs, a PI or colleague already has a GEO profile under their account—so the lab might say “GEO is on another account.”
  • Your options:

    • Create/use your own My NCBI + GEO Submitter Profile and submit under your name.
    • Or submit using the lab/PI account that already has the GEO profile (for consistent lab identity in GEO).

5) Practical “Data availability” logic for a manuscript

A journal-friendly setup is:

  • Raw sequencing reads (bulk RNA-seq FASTQs; later GeoMx FASTQs if available) → SRA accession(s)
  • Processed expression outputs and metadata (GeoMx DCC/PKC + annotation + processed tables) → GEO accession(s)
  • Workflow/code → GitHub + DOI archive (e.g., Zenodo), linked from GEO and/or the paper.

6) ENA note (from the email context)

  • If a manuscript currently lists an ENA project accession that can’t be found (e.g., looks like a placeholder), you either:

    • confirm the correct existing ENA project accession, or
    • create a new submission/accession (ENA or NCBI—both are widely accepted, but your processed GeoMx outputs still fit best in GEO).

7) Key takeaway (one-liner)

  • FASTQ = SRA (via NCBI Submission Portal + BioProject/BioSample)
  • GeoMx DCC/PKC + annotation + processed outputs = GEO
  • One My NCBI login, but GEO needs a GEO Submitter Profile and uses a separate submission entry point.

Leave a Reply

Your email address will not be published. Required fields are marked *