I wasn't sure if there was an easy way to do this so used the cat command to concatonate the data into a single forward and single reverse fastq.gz files (each sample now has 2 files). These are in folder Concatonated-sequences.
I tried to import these using the following script but got an error.
My best recomendation is to use a manifest format rather than the Casava format. Pro is that it gives you want more control. Con is that you have to build the manifest file. (Not so bad in your case, I suspect).
Do you know what you plan to do wtih the data after you get it loaded? Denoising, OTU picking? Something else? If you're planning to run Dada2, you should do that on a per-sequencing run level, but I think there's a weird scaling thing around depth and sample size. If you're doing open ref or de novo OTUs, you should import all the data and cluster together. Deblur doesn't really care.
Thankyou for your feedback. I have used the manifest format as you advised and was able to import my data. I have produced a summary which shows the number of samples, sequences and interactive quality, so I think I’m on the right track.
I was planning to denoise, and work my way through to alpha and beta diversity and hopefully taxonomy so I can compare the different samples. I will take a look at using deblur for denoising and go from their.