adaptor, barcode, and tag

[morning latte]

I am sorry for a very silly question. I am new to this field and confused with three terms which are adaptor, barcode, and tag. I am planning on sending my samples for illumina sequencing. Can anyone help me be clear with these three especially in terms of illumina? Thanks.

[SNPsaurus]

An adaptor is a short piece of known DNA attached to the genomic DNA you are sequencing (or cDNA if doing RNA-Seq). The Illumina adaptors contain the sequences needed to amplify the genomic fragment and hybridize to the sequencing adaptor.

Adaptors also contain short sequences needed to identify the sample, since many samples can be mixed together and sequenced on a single lane. Here the language gets a little fuzzy. A barcode, tag, MID, or index are all ways to describe the identifier. "Index" seems to usually refer to an identifying sequence that is read in by a sequencing read separate from the main read that sequences the genomic DNA. Barcode more usually refers to a short sequence that is read in the same read as the genomic DNA. There are just different ways to accomplish the same thing.

[JackieBadger]

well explained.

Just to add one thing; barcode/mid/tag are usually part of a modified primer. So when you PCR your produce it is incorporated into the sequence (thus you can have a fwd and rev barcode/mid/tag. Illumina's indices are not part of the sequence, but if you were to use indices to demultiplex on top of two barcodes, you get more demultiplexing power

[Heisman]

To add one last thing, whereas I've always used "index" and "barcode" as SNPsaurus described, some people don't do this so be crystal clear when discussing with people what definition of each term they are using.

[gene_x]

To add one more thing, in our sequencing service, barcode is added 5' of the sequence to be sequenced so it occupies some good quality bases at the beginning. Index, on the other hand, is at the 3' of the sequence to be sequenced. So I think index is preferred.

[Heisman]

To add one more thing, in our sequencing service, barcode is added 5' of the sequence to be sequenced so it occupies some good quality bases at the beginning. Index, on the other hand, is at the 3' of the sequence to be sequenced. So I think index is preferred.

Hmm, index is typically a separate read entirely; the data you get from the sequencing core may use a pipeline that puts the index onto the 3' end of the sequence, but it was actually a separate read (most likely).

[gene_x]

Hmm, index is typically a separate read entirely; the data you get from the sequencing core may use a pipeline that puts the index onto the 3' end of the sequence, but it was actually a separate read (most likely).

Right......

[Heisman]

Right......

Just clarifying it wasn't actually sequenced at the 3' end. If anything that would be a bad idea in the sense that the lowest quality bases are usually at the 3' end and that would give you the lowest recovery when trying to demultiplex your reads.

[gene_x]

Just clarifying it wasn't actually sequenced at the 3' end. If anything that would be a bad idea in the sense that the lowest quality bases are usually at the 3' end and that would give you the lowest recovery when trying to demultiplex your reads.

Yeah.That's right.

[SNPsaurus]

You would also need to have pretty exact insert sizes as well to ensure your read runs into a 3' barcode and doesn't waste a bunch of sequence on adapter.

One other definition. When we were working on developing RAD (microarray) and RAD-Seq, we used "tag" to refer to the genomic fragments flanking a cut site, since it "tagged" the cut site. Same with nextRAD, the library consists of "tags" at particular sequences. We wanted to not refer to every sequenced fragment as a marker, since many would not contain any genetic variation between the samples. So we used tag as a neutral term. But probably a little confusing!