RADCamp Latin America 2026 Part II (Bioinformatics) - Day 1 (PM)

Overview of the afternoon activities:

Prepare barcodes files

Take this time to create the barcodes files you will need for demultiplexing your raw data to samples (if this isn’t already done). The form of barcodes files is a tab separated text file, formatted with the first column being sample ID (no white space in sample ID names) and the second and third columns being the inner barcode sequences:

Sample1	CCGAAT   	CTAACG
Sample2	TTAGGCA	    CTAACG
Sample3	AACTCGTC	CTAACG
Sample4	GGTCTACGT	CTAACG
Sample5	GATACC		CTAACG
Sample6	AGCGTTG	    CTAACG
Sample7	CTGCAACT	CTAACG
Sample8	TCATGGTCA	CTAACG
Sample9	CCGAAT  	TCGGTAC

When you have your barcode file ready please upload a copy to the shared google drive for Participant Barcode Files.

Prepare other metadata files as needed

It could also be useful if you have other metadata to create a metadata.tsv file mapping sample IDs to whatever metadata you have, like sample site or region or morphological data, or latlongs, or whatever else you have. It should be formatted similar to the pops file, with the first column being sample ID and remaining columns containing metadata info. For the metadata file it will be useful to have the first row be reserved for column headers (no spaces in column header names, use underscore (_), like this:

SampleID    latitude    longitude   tarsus_length
1A          xxx         yyy         zzz

Finding reference sequences and/or alternate RADSeq data

In this activity, folks who don’t have data will search online for some interesting and relevant RADSeq data to download and re-assemble. For people with data you will search online for a relevant reference sequence to use during part of the assembly.

Finding RADSeq datasets

Look on sra and when you find a dataset, locate the BioProject number which looks like this PRJNA1499259. With the PRJ number follow the instructions on the ipyrad2 docs to Download the dataset.

Finding reference sequences

In ipyrad2 there are two modes of assembly: 1) denovo where loci shared among samples are identified without need of a reference sequence; and 2) reference-based where an external reference sequence can be used to facilitate the RADSeq assembly process. Usually we will recommend people to run both denovo and reference assemblies and compare them. If they agree then that’s great and if they don’t agree this can be something to dig into a bit further to better understand your data and the system.

In this workshop we will try to do both denovo and reference assemblies, so for this activity we will spend some time searching for reference sequences that are appropriate for each participant’s system.

Fill the reference sequence tracking google sheet with information about the taxa of your samples and a link to the most appropriate reference sequence you can find.

Intro to ipyrad2 subcommand mode

Walk through the subcommand mode here particulary for a reference based WGS + RADSeq dataset (perhaps using Amaranthus data).

Break for dinner and bowling activity

We will meet at Bolerama (across the street from Tec) for an early dinner and bowling/networking/social activity. This is an optional activity and costs will not be covered by RADCamp (unfortunately).