Cross-contamination between samples?

[ami...@gmail.com]

Amazing paper guys, I enjoyed reading your BioRxiv submission.

With a computational background, I have a rather basic question regarding cross-contamination between samples. 

Similar to performing a qPCR in a "not-so-careful-lab", it seems to be difficult to get a 'negative' measurement (i.e. a "no infection found" state), simply because the templates are floating around the lab if this is done frequently in such a scale. I was wondering with the barcoding system implemented in the LAMP-seq assay, what would happen if you get contamination of one LAMP reaction (one patient with positive infection) into another tube that is then used for a new LAMP reaction (healthy individual)?

 

Will you be able to see that this was a contamination because the sample contains the wrong barcode, or mixed barcodes? If yes, what this will look like in a sequenced dataset?

Thanks!

[Jonathan Schmid-Burgk]

Thanks for this thoughtful point! I would expect that contamination of an RT-LAMP reaction with the reaction product of another LAMP reaction, even when having a different barcode, will lead to false-positives that cannot be told apart by sequencing data. Therefore, it will be crucial that no LAMP reaction tube ever be opened at the site where fresh swabs are taken.