RADCamp Latin America 2026 Part II (Bioinformatics)

Day 2 (AM)

Overview of the activities:

Demultiplex Empirical Data

Lead: Isaac (Could take all morning)

Setting meaningful sample names in the barcodes file

Lead: Deren

Barcodes files in the google drive

The sample names in the barcodes files are going to follow the sample data all the way down-stream into the analysis. It is far easier to give the samples meaningful names at the very beginning (in the barcodes files) than it is to change sample names downstream (which is possible but can be annoying). Let’s take some time this morning to make sure our barcodes files have meaningful sample names.

Running ipyrad2-classic Step 1

The first step in a RADSeq assembly is to evaluate the quality of the raw data. It is typically more useful to do quality control on a per-sample basis rather than on the full undemultiplexed raw data file, so before we run fastqc we need to run ipyrad2 step 1 to demux raw data to samples.

Make a new directory and a new ipyrad2 params file for your empirical assembly

cd ~
mkdir <your_assembly_name>
cd <your_assembly_name>
ipyrad2-classic -n <your_assembly_name>

Edit params file to point to raw fastq data and barcodes file

Edit your new params file and set raw_fastq_path and barcodes_path to the full path where these files are found.

Run step 1 to demultiplex your data

ipyrad2-classic -p params-<your_assembly_name> -s 1 -c 16

Quick check of demux process

Demux fastq files are stored in the *_fastqs directory, so take a look in this directory to get a quick feel for how the data is distributed among samples:

# `-l` gives a detailed (long) list
# `-h` shows file sizes in human readable format 
ls -lh <your_assmbly_name>_fastqs

Question: Do most of your R1/R2 files for each sample show about the same size? Or are some much bigger or much smaller than average?

TODO FINISH

Empirical Data QC

Lead: Isaac (45’)

Now we will run fastqc on a couple of samples from the empirical data. To do this we will first change directory (cd) to where the demultiplexed samples live, inside ~/empirical/*_fastqs. Consult the fastqc exercise from yesterday for how to run and interpret results.

In the remaining time choose 1-2 samples to run fastqc on both R1 and R2. After the runs finish, inspect the results and try to come to a conclusion about what the results indicate.

Be prepared to answer the following questions:

Coffee break

Briefly report back on fastqc results

Empirical assembly

Create the imap file for selecting samples for Step 3

We are going to take a little time to carefully create an step3_samples.txt file for each participant to ensure Step 3 (denovo) will run efficiently. Create a new file called step3_samples.txt and add 5-7 samples that are representative of your data, one sample ID per line. Ask Deren and Isaac to check your work before proceeding.

You already started your empirical assembly when you ran Step 1 to demux your raw data to samples, so now you can continue with steps 2-5 to complete the assembly.