Further DNA Sequencing and Trials and Tribulations
Introduction:
Many previous weeks of work has been dedicated to developing DNA extraction protocols for third generation sequencing in order to produce data that will give insight to Deinoccocus species that have been neglected by the scientific community thus far. Last summer the method of choice after lengthy testing and the urgency to use a protocol for high molecular weight (QIAGEN Genomic tip protocol) was decided to be sonication for the lysing of cells because according to previous gels these samples produced high molecular weight and visible bands greater than 45kb, the highest seen in this lab to date. This week, we attempted to sequence two different species. The first. Deinococcus sonorensis was originally isolated by sonication and the lysate directly cleaned with the ZYMO DNA cleanup and concentration kit, which did not consistently work for other species after these runs. The second specie, Deinococcus pimensis was isolated as described from last week, with sonication and genomic tip which came out as nearly a perfect sample.
Methods:
The first species tested was Deinococcus sonorensis and the previously prepared sample in PCR water was retested on the Nanodrop and the fluorometer to confirm cleanliness and concentration. It was then diluted to a workable concentration because the protocol total mass input was between 100 and 150ng. This dilution was again confirmed and the volume of maximum 10 ul was added and normalized to 10 ul. Then incubated for 2 minutes at 30 and 2 more minutes at 80 degrees Celsius after the addition of fragmentation solution and mixed thoroughly by pipette. A sequencing prep solution was added to the sample and another 5 minute incubation at RT after being mixed thoroughly again, and the sample was ready for sequencing. Using the MinION sequencer with a flongal as opposed to an entire flow cell, the sequence was loaded and ran. It was seen that although the flongal had passed preliminary checks to ensure pore availability and strength, only one pore was active and intermittently sequencing. It was decided to redo the sample again in an identical fashion and attempt to sequence again, but to no avail again besides a few minutes worth of data.
Deinococcus pimensis was then prepared the same way another day to be loaded onto another flongal. This flongal also passed the preliminary checks, and the D. pimensis sample was arguably the best produced from all the species thus far in terms of cleanliness and being eluted in very clean water. That being said, the first flongal again did not work with the exception of a few pores receiving data, and so a second flow cell was attempted but it didn't even pass the initial check. Finally, a fifth flongal was used to confirm with even using specialty reagents in a different set that the likely issue is actually not the sample at all. And as expected, the fifth flongal also did not prove to be effective. It was decided that the company must be contacted for replacement of the flongals and reagents and sequencing would commence once this situation was resolved.
Results:
Although no usable data was retrieved from the sequencing attempts this week, one aspect of data was able to be found which indicated that the molecular weight of these samples were very poor in terms of length. This was the opposite of what was to be expected, in which all of these samples and the Deinococcus aquaticus isolates previously attempted showed an average N50 value (average base call number per fragment read) was only about 500 to 600 bp in length, compared to the Deinococcus aquaticus sequence from last year in which the average was roughly 2000 kb after being bead beated.
Discussion:
This week may have seemed like a waste in terms of the fact that no usable data was retrieved for data analysis, but it did give us the potential indication that sonication of samples is actually a detrimental process for the high molecular weight of the DNA input for sequencing. Even though we were confident that high molecular weight was being achieved, it may be possible that our gel and gel dye is just not sensitive enough to accurately show how much shearing is taking place through this process. It will be determined if it may be more worthwhile to go back to beadbeating or experimenting with other chemical lysis methods to prevent this shearing effect, because it will inevitably cause the bioinformatic analysis post sequencing to make sense of the data will be much more difficult. It may even cause confusion with the number of contigs for a particular species, and blur how accurate the sequence is, which would be catastrophic for potential publishing.
Comments
Post a Comment