Does anyone know the method of Moleculo's Long DNA Fragment Sequence?

[silin284]

Illumina's Moleculo Provides Long DNA Fragment Sequences to Improve Genomes, Clinical Assays

January 15, 2013

By Julia Karow

This article, originally published Jan. 14, has been updated with additional information from Illumina.

A week after Illumina acquired San Francisco-based startup Moleculo for an undisclosed amount, some early-access customers spoke about their experience with the firm's long fragment sequences for the assembly of complex genomes.

Moleculo, which has developed a technology to break genomic DNA into large pieces, sequence them with short reads, and assemble these into so-called "long reads," said it sees applications in the de novo assembly of plant and other complex genomes, human genome sequencing, and cancer sequencing.

****EDITED FOR PAID CONTENT****

Please see here for the full article: http://www.genomeweb.com/sequencing/...nomes-clinical

[ferlandl]

I haven't seen any datasets or other details about the technology but from my reading of this and other upper level articles, it seems to me that "long reads" is a misnomer for what Moleculo provides. There is a word for the result of stitching together the reads provided by sequencing (with any technology): it's called a contig. Granted, since they claim to be able to provide haplotyping information, suggesting some sort of individual fragment information, this appears to be more (and better) than just shotgun assembly. So maybe these are more than contigs, but they are still not "long reads".

Pending more details about the technology, may I suggest the new term "haplotig" for these constructs? That is, if my understanding from reading between the lines of these articles is correct.

[scbaker]

I think what Moleculo is doing is creating a bunch of mini-libraries from long (10-15kb) sequences. Sequences from each mini-library are connected via a unique sequence tag so that the original long sequence can be reassembled informatically from a sequence run of multiple mini-libraries.

[genbio64]

I think what Moleculo is doing is creating a bunch of mini-libraries from long (10-15kb) sequences. Sequences from each mini-library are connected via a unique sequence tag so that the original long sequence can be reassembled informatically from a sequence run of multiple mini-libraries.

from what I have been told its kind of a hybrid optical mapping/mate pair strategy where they cut the reads up at specific points, add specific primer pairs across the length, sequence and then stitch them together.

[ferlandl]

I think what Moleculo is doing is creating a bunch of mini-libraries from long (10-15kb) sequences. Sequences from each mini-library are connected via a unique sequence tag so that the original long sequence can be reassembled informatically from a sequence run of multiple mini-libraries.

from what I have been told its kind of a hybrid optical mapping/mate pair strategy where they cut the reads up at specific points, add specific primer pairs across the length, sequence and then stitch them together.

But each library would have to be a single molecule library. Various mate pair technologies would easily allow to isolate the molecules of interest from the rubbish, e.g. via a circularization adaptor, but to isolate them from one another is less obvious; one would still have to use a different adaptor for each circle and then use that to separate them (how?). Or, perhaps, just separate them by dilution? Perhaps this is where the optical mapping part comes in. Whatever they came up with, it looks pretty ingenious.

[scbaker]

Or, perhaps, just separate them by dilution?

I'm assuming dilution is involved, similar to Complete Genomics' LFR technology, but perhaps with microfluidics.

[ECO]

Edited the first post by request. Please don't paste full (*paid*) In Sequence articles here, especially without linking the source. They are good people that work hard to produce their content.

Thanks!

[tbanks]

I'm assuming dilution is involved, similar to Complete Genomics' LFR technology, but perhaps with microfluidics.

I spoke to some Illumina reps at PAG and they said kits were coming in late 2013 and that no specialized equipment will be needed. I mentioned microfluidics/Rain Dance and was told "nope...molecular biology". This being such a new development the reps could be misinformed but I am hoping for something really clever like the Nextera method was.

[krobison]

from what I have been told its kind of a hybrid optical mapping/mate pair strategy where they cut the reads up at specific points, add specific primer pairs across the length, sequence and then stitch them together.

No, no reason to think there is optical mapping involved. The libraries they make for you are standard Illumina libraries, dual barcoded, and can go on a flowcell with other libraries in a standard process.

Each "long read" (I have used the term "synthetic read"; I believe the Shendure paper along these lines used the term "subassembly") is derived from a single size-selected fragment.

[JPC]

I've been told there are two ways of doing this but I've only figure out one. With the Nextera protocol a tag is insterted which creates a break in the DNA and tags both ends. If that inserted tag contains a barcode which applies to the two now separated ends then it can be found in software and the two reads can be re-assembled. This barcode would need to be some random sequence, a palindrome would work great.

The sub sampling of the completed library to fill the flow cell means that not all fragments can be paired in this way limiting the size of the 'synthetic reads'.

[leewatkinsjr]

An interview in Bio-IT World with co-inventor Mickey Kertesz (who is/was a postdoc in Stephen Quake's lab) says there are related patent filings: http://www.bio-itworld.com/2013/1/18...quisition.html "The core technology is patent [pending] by Stanford and we in-licensed it to the company. [Steve Quake is also a co-inventor.] We filed additional patents during this year." So maybe we could find these patent filings and stop speculating on how it actually works.

[ferlandl]

(I have used the term "synthetic read"...)

I like that!

[Nitrogen-DNE-sulfer]

Its just happy mapping with nextera.

dilute DNA to LMOI (5% of genome per well)...Make Nextera Libraries with specific barcode pair per well. All fragments with that barcode can be assumed to be from a haploid copy do to the dilution they are at. Since Nextera leaves a 9bp gap upon integration, this gap is a signature on how to stitch synthetic reads together based on start point information.

[GW_OK]

http://www.illumina.com/technology/moleculo-technology.ilmn

Yep.
Tagging clonally amplified 10kb stretches. Sub-library making, I suppose you could call it. Not sure how they're amping up the 10kb fragments, though.

[ScottC]

... so how many unique synthetic reads do you get?