Effect of a 2.5 hour power outage on a SOLiD run.

[pmiguel]

Power was lost to several Purdue buildings on Saturday for 2.5 hours, including one housing a running SOLiD. By the time I arrived on site, power had been restored for about 1.5 hours. The instrument and its computer blades were all back up. After restarting ICS, priming the lines and resuming the run, it continued.

We probably got lucky as to when the power went out (just after a ligation completed). Also there was only one flow cell occupied. Nevertheless, near as we can tell at this juncture, the run will be okay.

Just one data point, but, fair is fair, I have to count this as a big win for Applied Biosystems.

--
Phillip

[colindaven]

That sounds promising. Can you please update this thread if something did go wrong which you haven't picked up yet ?

[pmiguel]

Sure. The run should complete by this weekend. Will let you know how it looks after mapping to a reference genome.

I should mention that the ligation prior to the power interruption was the 10th one for that primer. So, after scanning completed for the slide, the nascent strand was washed away ("reset") in preparation for the next primer.

Hence, our major concerns are somewhat limited. Of primary concern was that a bubble formed in the flow cell that allowed the beads to dry out. I pulled the flow cell down to take a look before resuming the run -- no obvious bubbles, although my ability to see a bubble through the layer of beads was uncertain. Then there is the possibility that after 2.5 hours without power, the reagents would degrade due to loss of chilling in the reagent block. No immediate problems there. Subsequent ligations look within spec. But some issues might be subtle...

--
Phillip

[pmiguel]

Okay, the mapping results are back.

Near as I can tell, the 2.5 hour power outage had no effect on the quality of the data subsequently collected. The outage occurred during second tag (F3) sequencing. The percentage of reads mapping to our reference was 71.1% for R3 (first tag) and 68.7% for F3 (second tag). Typically we see a 3-10% drop in % of tags mapping between the first and second tags.

Just to be clear, had the power outage happened at another stage, the ligation might have been spoiled, necessitating (at least) our re-running that primer.

But it does appear that in this instance the instrument weathered an unplanned 2.5 hour power outage prior to my resuming the run without detectable data loss. Although I came on site to resume the run, I could have done so remotely.

--
Phillip

[VanessaS]

We had a similar occurrence. When I saw the bubble, I panicked thinking everything was ruined! Apparently, according to my FAS, the flowcell maintains moisture and humidity so even though a large bubble may freak you out, most the times the beads are fine.

[pmiguel]

Was this with v4 (HD deposition)?

I ask because we had a case where our storage buffer valve was clogged and this somehow lead to many bubbles being pulled into the flow cell during the change between R3 and F3 tags for a mate pair run.

We lost the whole run because after this all the beads fell off. So I thought this might mean that the new deposition chemistry was much more sensitive to exposure to air.

--
Phillip

[VanessaS]

Yes this was with v4 chem. Our syringe bit the dust......

[pmiguel]

Well that is good to hear. (That the bubble did not hurt your run, not that your syringe died.)

Probably it was the air being pulled in through the side gaskets that created sufficient shear forces to blow all the beads off. Applied Biosystems should provide psychological counseling for customers who have lost a run...

--
Phillip

[VanessaS]

Well that is good to hear. (That the bubble did not hurt your run, not that your syringe died.)

Probably it was the air being pulled in through the side gaskets that created sufficient shear forces to blow all the beads off. Applied Biosystems should provide psychological counseling for customers who have lost a run...

--
Phillip

Amen Brotha, testify! This run... was actually a re-run from when the flowcell on the other instrument died. On the same samples. Twice.( it didnt really die, turns out it was the thingamajig that controls it) So having the syringe explode splendidly was emotionally damaging. We yanked the slide off of the bad flowcell, threw it on the second instrument just to have something else go boom. We fondly refer to that run as the Cursed Run from Hades. After the syringe was replaced there was a bit of glass or something that caused similar symptoms, but my AB engineers rock the house and they fixed, my FAS soothed and I only broke into tears once.