In addition to my earlier post on recovering data from a stopped MiSeq run, I want to share a method of rescuing the entire run. Basically, if MiSeq quits on you during read 1 phase (SE or PE run), here is a chance to get around the problem and salvage the kit.
1. Find the last cycle number, analyze intensity and quality plots and decide at what number you are comfortable to stop the run
2. Get RFID bypass code from ILMN. It is good for a week.
3. Start pos-trun wash as suggested by the software. If it is not suggested, quit the control software and restart it, then perform wash in maintenance. It is a good idea to restart software anyway as hardware needs to be initialized and everything homed after run was aborted.
4. Pull out the reagent cartridge and the PR2 bottle. Pluck out the RFID chips from both.
5. Aspirate contents of wells 5 (AMS1), 9 (LMX1), 17 (your library), wash with water, refill with water.
6. Add more read 1 primer (custom or standard) diluted as recommended in the manual, refill water wells.
7. If post-run wash is complete, pull out the flow cell. Carefully remove plastic cover, pluck out the RFID chip, blot excess water and snap in the cover back.
8. Edit the sample sheet (using either Excel or text editor) of the failed run - modify the reagent kit barcode (I just increment it by 1), number of cycles (SE or PE) as decided earlier and save under the name of the modified kit barcode.
9. Start new sequencing run. If it complains about unfinished analysis of the previous run, terminate analysis - use the procedure I described to get data out.
10. Enter new barcodes for the flow cell, the reagent kit (matching that entered in the modified sample sheet) and PR2 using the RFID bypass code. I just slightly modify the originals, like adding 1 somewhere. Make sure you specify the correct sample sheet.
11. Hit the run.
Basically the idea was to mock the cluster generation phase and start run 1 anew. In all cases I tried so far, the procedure worked both for SE and PE runs. The only problem is that, while most reagents remain in sufficient quantities, LPM (well 7) and RMF (well 11) may dwindle below sufficient. Any ideas what those could be? I guess the only way, until their composition becomes known, is to save excess of those from used reagent kits. Keep in mind though, that v1 and v2 may not be compatible as different abbreviations are used for some reagents.
1. Find the last cycle number, analyze intensity and quality plots and decide at what number you are comfortable to stop the run
2. Get RFID bypass code from ILMN. It is good for a week.
3. Start pos-trun wash as suggested by the software. If it is not suggested, quit the control software and restart it, then perform wash in maintenance. It is a good idea to restart software anyway as hardware needs to be initialized and everything homed after run was aborted.
4. Pull out the reagent cartridge and the PR2 bottle. Pluck out the RFID chips from both.
5. Aspirate contents of wells 5 (AMS1), 9 (LMX1), 17 (your library), wash with water, refill with water.
6. Add more read 1 primer (custom or standard) diluted as recommended in the manual, refill water wells.
7. If post-run wash is complete, pull out the flow cell. Carefully remove plastic cover, pluck out the RFID chip, blot excess water and snap in the cover back.
8. Edit the sample sheet (using either Excel or text editor) of the failed run - modify the reagent kit barcode (I just increment it by 1), number of cycles (SE or PE) as decided earlier and save under the name of the modified kit barcode.
9. Start new sequencing run. If it complains about unfinished analysis of the previous run, terminate analysis - use the procedure I described to get data out.
10. Enter new barcodes for the flow cell, the reagent kit (matching that entered in the modified sample sheet) and PR2 using the RFID bypass code. I just slightly modify the originals, like adding 1 somewhere. Make sure you specify the correct sample sheet.
11. Hit the run.
Basically the idea was to mock the cluster generation phase and start run 1 anew. In all cases I tried so far, the procedure worked both for SE and PE runs. The only problem is that, while most reagents remain in sufficient quantities, LPM (well 7) and RMF (well 11) may dwindle below sufficient. Any ideas what those could be? I guess the only way, until their composition becomes known, is to save excess of those from used reagent kits. Keep in mind though, that v1 and v2 may not be compatible as different abbreviations are used for some reagents.
Comment