SEQanswers

Go Back   SEQanswers > Sequencing Technologies/Companies > SOLiD



Similar Threads
Thread Thread Starter Forum Replies Last Post
How good bowtie can map SOLiD data? ypsd SOLiD 11 12-17-2012 07:56 AM
SOLiD reads map using BFAST and Samtools jsun529 SOLiD 17 02-24-2011 12:55 PM
How to map SOLiD paired end reads by Bfast beliefbio Bioinformatics 1 12-29-2010 12:55 AM
How to map NCBI SOLiD reads with BWA Chandana Bioinformatics 3 12-04-2010 05:57 AM
map solid reads to ncRNA database crh SOLiD 2 07-19-2010 11:22 AM

Reply
 
Thread Tools
Old 09-07-2010, 07:38 AM   #1
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,307
Thumbs up Repeating a SOLiD Focal Map

If you are like me, the very idea of repeating the focal map of a run after the run has finished will fill you with a certain amount of horror. This must be informatics equivalent of replacing the foundation of a skyscraper after the building is complete. But it turns out it is not only possible, but fairly trivial to do -- with a major caveat: reprocessing the entire run, from image analysis on, then occurs and this requires the better part of 5 days! But hey, if nothing else, your SOLiD will, by now, have taught you patience.

For reasons unknown, but probably the result of the dew point spiking above 15 oC in our lab, the focal map for our most recent run had a large number (>5%) of blurry panels.

In typical AB fashion the run proceeded on nonetheless, oblivious to the issue. Well, perhaps "oblivious" is unfair. How about "robust to the failure to collect a number of focal map panels that would prevent subsequent high quality data collection from those self-same panels from being utilized?" Perhaps the instrument was stoically soldiering on? One would hesitate to characterize our SOLiD as the "strong silent type". Rather, the dramatics to which we are continually subjected have a distinctly passive-aggressive feel to them.

In this case we only became aware of the issue because recent climate-control issues have caused us to keep our eyes glued to the heat maps of every cycle of every run. Helpfully, the instrument software (ICS) has a heat map client that actually allows a single panel to be double-clicked to display the original focal map adjacent to the 4 filter-wheeled images collected in that ligation cycle. All the "black" panels had great looking images -- except for the focal map itself, that was very blurry in every case where a panel had failed.

We were advised by Applied Biosystems to repeat the focal map as soon as possible. But we were already into the F5 reads at that point. Those working with AB Paired End chemistry will be aware that it is well-nigh impossible to get hold of currently. So we did not want to risk those reads. Hence, we waited another week to let the entire run finish, copied the resulting base calls, quals etc. off and only then repeated the focal map.

Okay. One surprising aspect this process was the ease with which this was accomplished. There were no instructions in the instrument manual for "repeat the focal map". But, basically, none were needed. Just put a focal map regent-containing strip from any kit (we used a WFA) in the right slot and set the run point back to the beginning of the run. Unpause the run. Set a pause point to prevent any actual ligations from being performed. That was it.

Without further prompting SETS drew the correct conclusion "Ah, the very basis of my calling every single bead, for every single ligation has been replaced. I shall get right to the task of repeating, from the ground up, every image analysis task I completed during the last two weeks."

So seemless was this process, that until I looked at "cluster view", it was not even clear to me that the gambit had succeeded. Here is a plot from ganglia for the system load over the past week:



Notice the normal ticks of heavy activity as each ligation completes at the beginning of the week. Then notice the end of the week and through the weekend -- cluster maxed out.

Meanwhile perusing cycle scans lets you know where SETS currently is. I could not imagine a more delightful implementation of what I would have to imagine would be a rarely used procedure. The best possible solution. Well, excepting, perhaps, including some $10 sensor inside the instrument that would warn me that the climate control in my lab was so woefully inadequate that part of the slide had fogged up and data collection should be paused until I have that taken care of. But that might be asking too much because I get the distinct impression that the relevant decision-making entities of Applied Biosystems cannot imagine a lab where the temps would wander above 75 oF when the humidity was above 50%. Or, even if, intellectually, they realize this is a possibility, the thought of actually pointing this out to a customer seems so irremediably rude that it simply cannot be countenanced.

--
Phillip
pmiguel is offline   Reply With Quote
Old 09-08-2010, 07:56 AM   #2
KevinLam
Senior Member
 
Location: SEA

Join Date: Nov 2009
Posts: 197
Default

Great Post!
It's nice to know there are people more willing to share knowledge than the ones you expect the knowledge to come from!
KevinLam is offline   Reply With Quote
Old 09-08-2010, 10:02 AM   #3
westerman
Rick Westerman
 
Location: Purdue University, Indiana, USA

Join Date: Jun 2008
Posts: 1,104
Default

Phillip is out sick today otherwise he would probably post about our post-rerun-focal-map results. We gained about 7% more reads. This is a barcoded run. Most of the gain was from reads going from 'unassigned' to one of the barcoded libraries. I can also see that there are more reads with 0 mismatches in the barcode and fewer with 1 mismatch in the barcode. Additionally the number of lines in the qual with the poor quality score of "-1" has decreased by about 40%

All of this indicates that not only did we recover more reads but we also increased the overall quality of the reads. The 5 extra days of processing time (especially of the USA's Labor day holiday) was probably worth it in this particular case. Now to run the data through the rest of the pipeline.
westerman is offline   Reply With Quote
Old 09-09-2010, 03:34 AM   #4
pmiguel
Senior Member
 
Location: Purdue University, West Lafayette, Indiana

Join Date: Aug 2008
Posts: 2,307
Default

Yes, Purdue is its normal early autumn hotbed of pestilence.

One of the bizarre things about the bad focal map was that those terribly blurred images did not result in black panels for every subsequent ligation. But my guess is that none of the reads from the blurry focal map panels was going to map back to a reference genome.

--
Phillip
pmiguel is offline   Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off




All times are GMT -8. The time now is 05:47 PM.


Powered by vBulletin® Version 3.8.9
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Single Sign On provided by vBSSO