Questions Illumina technology

Juergen F · 01-27-2012, 12:15 PM

Hi,

Got some question regarding their technology?

1 How/ why do the prepared libraries spread equally distributed over the space of the flow cell, and do not piling up/ condensing at the entrance?

2 How does a “cluster defend it’s space” in a flow cell and interface with the next cluster/ starting molecuele?
… 2a) …during initial binding
… 2b) … during isothermal bridge PCR

3 How do I know/ control that “I do not overfill” a flow cell, assuming that # of potential clusters is limited?

4 How does sample loading work on cBOT, does the machine have a sample input tray

5 Doing indexed sequencing I need to:
- do library preparation individually per sample, and pool aftrewards the material (all sample libraries) prior a cBOT run into one flow cell

OR

- do I prepare the library individually per sample and feed sample material sequentially into the cBOT process

6 Does somebody have process description for cBOT and MiSeq (what is the instrument doing step by step: pipette 1st sample …. Incubate … pipette reagent X …)

Thanks & Greetings Juergen

krobison · 01-27-2012, 07:31 PM

Quote:

Originally Posted by Juergen F

Hi,

Got some question regarding their technology?

1 How/ why do the prepared libraries spread equally distributed over the space of the flow cell, and do not piling up/ condensing at the entrance?

2 How does a “cluster defend it’s space” in a flow cell and interface with the next cluster/ starting molecuele?
… 2a) …during initial binding
… 2b) … during isothermal bridge PCR

3 How do I know/ control that “I do not overfill” a flow cell, assuming that # of potential clusters is limited?

Thanks & Greetings Juergen

I think the answer to #1 is simply turbulence during loading; enough molecules shoot past. I haven't seen a similar plot for other platforms, but on Ion Torrent you do definitely see patterns in the loading emanating from the input port

For #2, I think the key concept is that the surface-linked primers are limiting. When a cluster forms, it quickly ties up all primers in close proximity to the initial molecule. Growth continues outwards until it bumps into the outward growth of another cluster. The clusters can't grow through each other, because the primers are exhausted within each cluster.

That somewhat answers #3: the more you load on, the smaller the clusters. Smaller clusters yield less signal, so overload the flow cell & get poor results.

gringer · 01-28-2012, 02:39 AM

Regarding 1, DNA will only anneal below a particular temperature. My first thought-experiment suggests that if the solution were homogenised and passed over a heated plate and then cooled, the distribution of clusters should be somewhat representative of the homogeneity of the solution.

Regarding 2, I know that it is possible to have a cluster density that is too high due to high sample loading concentration, so it is possible that there are no checks in place to prevent overloading. My second though experiment suggests it would be possible to protect against overloading by starting the surface off with all but a few specific sites pre-annealed. After the initial samples are loaded, a reaction is carried out to free up only the protected sites for subsequent bridge amplification. I very much doubt this is what Illumina does, because if it were done you would already know what parts of the surface were potential sequencing sites.

HESmith · 01-28-2012, 07:47 AM

1) Typically, there is a decreasing gradient of cluster density from the intake to outflow side.
2a) Clustering is random and concentration-dependent.
3) There is no limit, so use the recommended loading concentrations (~8-10 picomolar).
4) Samples are loaded in a 8-tube strip.
5) Libraries are prepared individually and pooled before loading into the tube strip.
6) The user guides are available from the Illumina website (you have to register first).

miss_shell · 05-01-2012, 07:12 AM

For question 4. The samples are loaded onto the C-bot in 8-tube strip as mentioned by HESmith above. A manifold is then used to transfer the samples onto the flow cell.
For questions 5. We use a 96 well plate from the shearing through to library prep. We then pool the samples accordingly into a tube.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Newbie questions regarding Illumina read quality statistics using FASTX toolkit	Lspoor	Bioinformatics	21	09-05-2013 12:48 PM
questions of illumina pe reads fastqc results	arrchi	Bioinformatics	1	12-01-2011 04:07 PM
Questions on the updated illumina quality score	zeam	Bioinformatics	6	10-26-2011 12:08 PM
Simple questions about Illumina	Liselotte	Illumina/Solexa	7	06-29-2011 05:26 AM
questions about Illumina mRNA sample prep	ik76	Illumina/Solexa	2	01-19-2010 04:38 AM

01-27-2012, 12:15 PM	#1
Juergen F Junior Member Location: California + Germany Join Date: Jan 2012 Posts: 3	Questions Illumina technology Hi, Got some question regarding their technology? 1 How/ why do the prepared libraries spread equally distributed over the space of the flow cell, and do not piling up/ condensing at the entrance? 2 How does a “cluster defend it’s space” in a flow cell and interface with the next cluster/ starting molecuele? … 2a) …during initial binding … 2b) … during isothermal bridge PCR 3 How do I know/ control that “I do not overfill” a flow cell, assuming that # of potential clusters is limited? 4 How does sample loading work on cBOT, does the machine have a sample input tray 5 Doing indexed sequencing I need to: - do library preparation individually per sample, and pool aftrewards the material (all sample libraries) prior a cBOT run into one flow cell OR - do I prepare the library individually per sample and feed sample material sequentially into the cBOT process 6 Does somebody have process description for cBOT and MiSeq (what is the instrument doing step by step: pipette 1st sample …. Incubate … pipette reagent X …) Thanks & Greetings Juergen

01-28-2012, 02:39 AM	#3
gringer David Eccles (gringer) Location: Wellington, New Zealand Join Date: May 2011 Posts: 838	Regarding 1, DNA will only anneal below a particular temperature. My first thought-experiment suggests that if the solution were homogenised and passed over a heated plate and then cooled, the distribution of clusters should be somewhat representative of the homogeneity of the solution. Regarding 2, I know that it is possible to have a cluster density that is too high due to high sample loading concentration, so it is possible that there are no checks in place to prevent overloading. My second though experiment suggests it would be possible to protect against overloading by starting the surface off with all but a few specific sites pre-annealed. After the initial samples are loaded, a reaction is carried out to free up only the protected sites for subsequent bridge amplification. I very much doubt this is what Illumina does, because if it were done you would already know what parts of the surface were potential sequencing sites.

01-28-2012, 07:47 AM	#4
HESmith Senior Member Location: Bethesda MD Join Date: Oct 2009 Posts: 509	1) Typically, there is a decreasing gradient of cluster density from the intake to outflow side. 2a) Clustering is random and concentration-dependent. 3) There is no limit, so use the recommended loading concentrations (~8-10 picomolar). 4) Samples are loaded in a 8-tube strip. 5) Libraries are prepared individually and pooled before loading into the tube strip. 6) The user guides are available from the Illumina website (you have to register first).

05-01-2012, 07:12 AM	#5
miss_shell Junior Member Location: cambridge Join Date: Mar 2012 Posts: 4	For question 4. The samples are loaded onto the C-bot in 8-tube strip as mentioned by HESmith above. A manifold is then used to transfer the samples onto the flow cell. For questions 5. We use a 96 well plate from the shearing through to library prep. We then pool the samples accordingly into a tube.