AI Blocks › Research a topic properly
16S microbiome read processor
Give it paired 16S reads and primers. Get a sequence variant table, species labels and a repeatable record.
16S microbiome read processor is a free AI skill for the job "Research a topic properly". Add it to Claude, ChatGPT, Gemini or Codex as a skill, or paste it into any chat.
Free · No account · MIT licence · Made by K-Dense · Level: expert
Made by K-Dense. See the original on GitHub; we copied it as it is, with its licence.
What it does
You end up with a table of sequence variants, a species label for each one, and a saved record of every step, so a colleague can rerun the work and get the same answer. The skill checks your setup before it runs anything. It confirms the quality encoding, the primer sequences as they were read, and whether primers are still on the reads. It uses your quality plots to set trim points, then tests that the forward and reverse reads will overlap by enough bases. After denoising it reports how many reads survived, and it helps you pick a classifier that fits your region. It needs QIIME 2 installed on your machine.
You give it
- Paired FASTQ files, one pair per sample
- Your primer sequences
- Sample metadata with matching IDs
- A trained classifier file for your 16S region
- Quality plots from a first look at the reads (optional)
You get
- A table of sequence variants per sample
- Species labels for each variant
- A read-retention report by sample
- Saved QIIME 2 files that record every step
- A short list of checks that passed or failed
Use it when
- You have paired 16S reads from a sequencer and need a clean variant table.
- A reviewer asked you to show exactly how you processed your microbiome data.
- You are not sure your trim settings leave enough overlap between read pairs.
- You need species labels for each variant using a reference you trust.
How to use it
- Add it to your AI once: download it and add the folder to Claude, ChatGPT or Codex as a skill.
- Ask for the job in your own words. Your AI opens the skill when it fits.
- It works best when your AI can run code.