# Human microbial samples taken with swabs - how many biological replicates are needed?
**URL:** https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626
**Category:** General Discussion
**Created:** [December 10, 2021, 9:46am UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626 "2021-12-10T09:46:27Z")
**Posts on this page:** 6
**Page:** 1
### Author:  [@Noemi\_Procopio](https://forum.qiime2.org/u/Noemi_Procopio)
#### Post date: [December 10, 2021, 9:46am UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626/1 "2021-12-10T09:46:28Z")
Hi all, I am about to extract human swabs taken from different anatomical locations (e.g., hand, neck, oral, etc) and I originally collected three biological samples (three swabs) per individual per location. I am now wondering if I have to extract them all (e.g., the three replicates) and perform sequencing of the 16S, or if I can extract only two replicates, look at the results, and eventually extract the third replicates if needed (e.g., if data amongst the two replicates are not in concordance)? Would this be an issue (e.g., sequencing in two different runs therefore using different chips), or this will be accepted? Also, will two replicates be enough or better to stay with three? Thank you!
---
### Author:  [@Nicholas\_Bokulich](https://forum.qiime2.org/u/Nicholas_Bokulich)
#### Post date: [December 13, 2021, 7:47am UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626/2 "2021-12-13T07:47:10Z")
Hi @Noemi_Procopio welcome to the forum!
I think this paper answers some of your questions:
> **[Triplicate PCR reactions for 16S rRNA gene amplicon sequencing are unnecessary...](https://pubmed.ncbi.nlm.nih.gov/31124709/)**
>
> Conventional wisdom holds that PCR amplification for sequencing should employ pooled replicate reactions to reduce bias due to jackpot effects and chimera formation. However, modern amplicon data analysis employs methods that may be less sensitive to...
> [@Noemi\_Procopio](#):
>
> Would this be an issue (e.g., sequencing in two different runs therefore using different chips), or this will be accepted?
Sequencing in different runs will likely lead to larger technical variations due to batch effects. So if you do want to sequence multiple replicate samples, do it all at once.
Good luck!
---
### Author:  [@Noemi\_Procopio](https://forum.qiime2.org/u/Noemi_Procopio)
#### Post date: [December 13, 2021, 8:40am UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626/3 "2021-12-13T08:40:20Z")
Dear Nicholas,
Thank you very much! I will definitely have a look at this paper! Thanks for your very useful replies 😉
Noemi
---
### Author:  [@Noemi\_Procopio](https://forum.qiime2.org/u/Noemi_Procopio)
#### Post date: [December 13, 2021, 9:31am UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626/4 "2021-12-13T09:31:57Z")
So, according to this paper it is useless to perform triplicate PCRs and pooling from a single sample. However, they do not specify if starting from multiple biological replicates (e.g., two-three samples taken from the same source/area, which result in two-three extractions prior to PCR) is something needed or not, and if so, how many (2 or 3)? Does anyone have any suggestion?
Thank you all again!
---
### Author:  [@jwdebelius](https://forum.qiime2.org/u/jwdebelius)
#### Post date: [December 13, 2021, 9:36am UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626/5 "2021-12-13T09:36:18Z")
My recommendation would be one. An individuals microbial signature tends to be quite strong. This adds a fun layer to your analysis where you need to use a model that accounts for individual effects and complicates the statistics. The only reason to extract two at once would be if you thought the sample might fail to amplify or if you plan to use the second replicate for another analysis (i.e. you're doing qPCR on the second swab).
Best,
Justine
---
### Author:  [@Noemi\_Procopio](https://forum.qiime2.org/u/Noemi_Procopio)
#### Post date: [December 13, 2021, 2:22pm UTC](https://forum.qiime2.org/t/human-microbial-samples-taken-with-swabs-how-many-biological-replicates-are-needed/21626/8 "2021-12-13T14:22:13Z")
Amazing, thanks for your reply!