FAQ

Q1. What is the iTru5_8N primer sequence?

Ans. Go to the Adapterama Protocols page and download the appropriate excel file. There are many tabs in these files, so please look at the different tabs.

Q2. What are the full sequences for the iTru primers and adaptors?

Ans. Go to the Adapterama Protocols page and download the appropriate excel file. There are many tabs in these files, so please look at the different tabs.

Q1. Why are you using so many types of RADseq/GBS?

Ans. We want to do lots of different things. Each flavor has specific circumstances when it is the best option.

Q2. Why do you have adapters in sets of 4, 8, or 12?

Ans. Illumina sequencers require base diversity at each base position to work correctly. We have designed the adapters to be well-balanced when used in sets of 4 (or multiples thereof).

Q3. Can I just purchase &/or use just one adapter?

Ans. No, not unless you: 1) mix in genomic libraries (e.g., PhiX) as a large fraction of the run (the genomic libraries create diversity in your run that is normally created by using the adapters in sets of 4) or 2) want to waste a lot of money & time. If you use only one adapter on either end & don't have anything else in your lane to create a reasonable amount of diversity, then the sequencing of that end will completely fail.

Q4. I just want to do a few samples, and I want to use a MiSeq. What should I do?

Ans. There are several options :

        1) Combine your samples with other things to create base diversity.

      2) Purchase 4 adapters (for each end), list your adapters on a 4x4 matrix and then do your ligations with adapters pairs on the diagonals; as few as 4 samples creates full adapter diversity with this approach.

        3) Make a pool of the >4 adapters (for each end) and rely upon unique i7 &/or i5 combinations for sample tagging.

Q5. I plan to use the i5 & i7 indexes to demultiplex all of my samples. Do I still have to use more than one adapter or mix in a genomic library?

Ans. YES (again, unless you like to waste time & money on things that are sure to fail). You need to either use multiple adapters, or mix in genomic libraries, or both. WE DO BOTH!

Q1.What is B-RAD?

Ans. Blunt-RADseq. B-RAD uses a single restriction enzyme and a single adapter (the same adapter is added to both ends of each piece of DNA).

Q2. When would you use B-RAD?

Ans. When you need a significant portion of the genome turned into a library where the ends are at consistent locations.

Q3. Why don't other people use B-RAD?

Ans. If you use a frequent cutting enzyme, then this does a poor job of making a library that is a small fraction of the genome (i.e., this makes a large representation library not a reduced representation library). So, it isn't accomplishing what most people want from a RADseq library.

Q4. Where did the idea for B-RAD adapters come from?

Ans. The general approach has been around for decades. TCG developed superSNX in the late 1990's, which derives from adapters that were developed years earlier. Our B-RADs are descendants of superSNX (sons of superSNX's, so to speak).

Q1.What is M-RAD?

Ans. Multi-RAD is a variant of RADseq where we add additional enzymes to the standard 3RAD approach.

Q2. Why would anyone use M-RAD?

Ans. If you want to efficiently reduce the number of loci being surveyed relative to 3RAD or any other dual-digest RADseq approach.

Q3. What are the advantages of M-RAD?

Ans. It should give a more repeatable small fraction of the genome than other approaches. For example, you might try taking a very small size-range from a dual-digest library preparation (3RAD or ddRAD or whatever), but this makes it difficult to get the same loci from all (or even most) samples. If you take a larger size-range, but reduce the number of loci within that range, that is easier to do consistently among samples (part of why RADcap works so well).

Q4. If you want to reduce the number of loci, why don't you just use less-frequent-cutting enzymes for 3RAD?

Ans. Although that is a straightforward alternative (and 3RAD is designed so a great number of enzyme combinations are feasible), in practice it is easier to add enzymes to 3RAD (thus converting 3RAD to M-RAD) than to fiddle with a bunch of different pairs of enzymes. It is also cheaper, because you have fewer enzymes to keep on hand & you can focus on using the cheapest possible enzymes. Finally, you can analyze the data you obtain from an initial 3RAD run to determine what additional enzymes you need to add to get to your target number of loci – so there is less guess work.

Q1.What are the differences between iNext and iTru library preparation

Ans. After polishing the sheared DNA (i.e. making the DNA blunt), iTru library preparations add an adenosine (A) monomer to each 3' end of the polished molecule. The iNext library preparation adds a cytosine (C) monomer to each 3' end of the polished molecule. Following addition of A or C, iTru or iNext specific adapters are ligated to the A/C tailed molecules. Finally, iTru or iNext specific primers are used. You can't mix and match. So, if you add an A to your inserts, you must then use iTru adapters and iTru primers. Alternatively, if you add a C to your inserts, you must then use iNext adapters and iNext primers.

Q2.Do the 5' prime ends of the A/C tailed molecules have phosphate groups attached?

Ans. Yes, the 5' prime ends of the DNA strands opposite the strand with 3' A/C do have phosphate groups (you are adding PNK as part of the polishing step to ensure phosphates on the 5' ends). This allows for ligation of the stubby Y-yoke adapters (ligated product).

Q3. Are the ends of the adapter molecule complementary?

Ans. No, the ends of the adapter molecule (the portion not directly ligated to the sheared DNA) are not complementary, and as such, the top and bottom strand do not anneal (shown as a separated, Y-shaped molecule). These two ends of the adapter have primer binding sites, to which the primers bind during PCR.

Q4. Which primers anneal in the first PCR cycle?

Ans. Only the iTru7 primers anneal in the first PCR cycle. The orientation of the denatured ligation product only allows for the iTru7 primer to anneal to the 3' ends. The iTru5 primer anneals during the second cycle.

Q5. How many cycles are necessary to produce the full length, double indexed libraries?

Ans. Three cycles are necessary to produce the full length, double-indexed library molecules.

Q6. Which portion of the library prepared molecule anneals to the flow cell?

Ans. The P5 and P7 ends anneal to complementary sequences present on the flow cell.

Q1. What is 3RAD?

Ans. 3RAD is a variant of dual-digest RADseq that uses 3 restriction enzymes, and makes use of simultaneous restriction digestion & ligation to efficiently add the adapters onto each piece of DNA with compatible ends.

Q2. What combination of enzymes should I use?

Ans. The only way to know is to determine it empirically via in silico testing (if you have a genome sequence) or in the lab (if you don't have a genome sequence). In practice, about 75% of taxa we try work well with design 1, and most the rest work OK with design 2. It is unusual to have to use other adapter designs (≤5% of projects we have done in the past ~ 5 years).

Q3. What are the advantages of 3RAD?

Ans. The main advantages of 3RAD include:

        1) An easy lab protocol with fewer steps.

        2) You need fewer oligos to make many combinations.

        3) You can pool 96 ligations together prior to doing any other manipulations.

        4) By using multiple adapters, each with varying length tags, base diversity is created within your library pools.

        5) These are quite efficient at getting linkers on the DNA. We have used as little as 0.05 ng of DNA as input (though we do NOT recommend using that little unless you modify the standard protocol and/or are OK with a random subsample of potential loci).

        6) You can use molecular identifiers to identify and remove PCR duplicates (see Hoffberg et al. 2016 for details).

Q4. What are the disadvantages of 3RAD?

Ans. The main disadvantages of 3RAD include:

        1) You get a variety of simple and complex products you don't want. Often, these are only a minor fraction of the reads, but sometimes they are a substantial fraction.

        2) It is more expensive to use the 3rd enzyme (i.e., the 3rd restriction enzyme costs money).

        The bottom line is that if you have plenty of input DNA, it is better to skip the 3rd enzyme. If your input DNA is sketch, then include it. We have not done both (skipping & using the 3rd enzyme within a project -- we either always use it or skip it for all samples).

Q5. What is the point of having the 3rd enzyme?

Ans. The third enzyme allows for read1 adapter dimers to be recut. This facilitates more product when input DNA is limited.

Q6. Do I have to use the 3rd enzyme?

Ans. No. You can skip the 3rd enzyme, which will make your life easier in several ways, but the disadvantage is that you can't get away with using only a tiny amount of template DNA. We have done some testing without the 3rd enzyme, and it works in practice (see 2RAD), but to date, we have more experience using the 3rd enzyme.

Q7. How do I demultiplex my data by internal barcodes?

Ans. Please see our Adapterama III manuscript for details. A link to the most recent version is on the Adapterama Publications page.

Q8. Why don't you worry about read2 adapters forming dimers

Ans. The read2 adapters are not phosphorylated, so they cannot self-ligate.

Q9. Why do you have dummy oligos on each adapter?

Ans. The dummy oligos make a double-stranded adapter that can be ligated with T4 DNA ligase. We only want a single strand of template to be formed, the other strand is discarded.

Q10. Why do you generate libraries from only a single strand of the template DNA?

Ans. The tricks we are using to get everything in the gamish to all work correctly simultaneously will not work if we try to make use of both strands. So, we are sacrificing one strand to gain efficiency in ligating onto the other strand.

Q11. At what cycle is the full PCR product generated?

Ans. The full PCR product is generated following the third cycle (there is a supplemental figure in the Adapterama III manuscript that shows this; you can also see this in on the Adapterama III video on YouTube).

Q12. Do both adapters (read1 and read2 ends) have phosphate groups at the 5' end?

Ans. No, only the Read1 adapter (illustrated on the left side) has a phosphate group at the 5' end.