# PICRUST2 Input Data setting
**URL:** https://forum.qiime2.org/t/picrust2-input-data-setting/33394
**Category:** Other Bioinformatics Tools
**Tags:** taxonomy, dada2, picrust
**Created:** [July 4, 2025, 4:37am UTC](https://forum.qiime2.org/t/picrust2-input-data-setting/33394 "2025-07-04T04:37:40Z")
**Posts on this page:** 2
**Page:** 1
### Author:  [@Youngtae\_Won](https://forum.qiime2.org/u/Youngtae_Won)
#### Post date: [July 4, 2025, 4:37am UTC](https://forum.qiime2.org/t/picrust2-input-data-setting/33394/1 "2025-07-04T04:37:40Z")
Hello, I am a researcher who is new to microbiome studies. I am reaching out to seek advice and insights from those with more experience regarding some questions I have about using PICRUST2.
I have been processing and analyzing fastq files with dada2, and I am now interested in performing additional functional prediction analysis using PICRUST2.
My main question concerns the input sequences for PICRUST2.
Should I provide all unique sequences from the original, unfiltered fastq files, or should I use the sequences that have been filtered and processed by dada2?
Additionally, I would like to know if there are any other tools, besides PICRUST2, that are useful for pathway or functional analysis, such as KEGG or GO analysis.
I am aware that I can find information on the QIIME2 forum, but as of posting this, I have not yet found the answers I am looking for.
I would greatly appreciate any advice or help from the community. Thank you very much.
---
### Author:  [@colinbrislawn](https://forum.qiime2.org/u/colinbrislawn)
#### Post date: [July 8, 2025, 12:04am UTC](https://forum.qiime2.org/t/picrust2-input-data-setting/33394/2 "2025-07-08T00:04:38Z")
> [@Youngtae\_Won](#):
>
> Should I provide all unique sequences from the original, unfiltered fastq files, or should I use the sequences that have been filtered and processed by dada2?
I would use the ASVs from DADA2 as discribed here: [q2 picrust2 Tutorial · picrust/picrust2 Wiki · GitHub](https://github.com/picrust/picrust2/wiki/q2-picrust2-Tutorial/)
> [@Youngtae\_Won](#):
>
> Additionally, I would like to know if there are any other tools, besides PICRUST2, that are useful for pathway or functional analysis, such as KEGG or GO analysis.
Yes! See Figure2 from the Picrust2 paper [PMC7365738](https://pmc.ncbi.nlm.nih.gov/articles/PMC7365738/)
- Piphillin
[GitHub - dmcskim/pyphillin: Functional profiling of 16S sequence data. Public, open source version of Piphillin.](https://github.com/dmcskim/pyphillin)
- PanFP
[GitHub - srjun/panfp](https://github.com/srjun/panfp)
- Tax4Fun2
[GitHub - fjossandon/Tax4Fun2](https://github.com/fjossandon/Tax4Fun2)
Remember that a 16S V4 amplicon contains only about 250 basepairs, which is not a lot of information, like only 500 bits or \<0.1 kB per ASV. It's miraculous that picrust2 gets more than halfway to shotgun metagenomic sequencing from so little data.
EDIT: My friend and colleague David Tool did his dissertation on functional prediction. View the PDF here: [https://ufdcimages.uflib.ufl.edu/UF/E0/05/61/89/00001/Toole\_D.pdf](https://ufdcimages.uflib.ufl.edu/UF/E0/05/61/89/00001/Toole_D.pdf)