Monthly Archives: August 2016

Synthetic fragment troubleshooting

A crucial step in the next steps in my thesis is to be able to synthesize a 200bp fragment with a randomize section. This fragment is planned to be synthesized by using two oligos, 135 bp oligo and a 95 bp oligo.


The strategy to synthesize the fragment includes anneal and extend the oligos and finally amplify the fragment using two 15-16 bases primers.


Annealing and extension steps show a 200bp band, as expected, with a lot of crude showing as a tail in the agarose gel.  When I used this anneal/extension product as template for a PCR, two bands are visible in an agarose gel as well as a lot of smear, a 200 bp band and a 250 bp band.





‘Anneal oligo’ showed the gel migration of a anneal only product, ‘extension only’ shows migration of anneal/extended product. 1 to 20 are products of PCR amplification using the anneal oligos as template using 1,5,10 and 20 cycles.






Looking at this result, my next strategy was to first gel extract the 200 bp band of the  anneal/extended oligo products, and I used it as template for a PCR. Results showed that now the 200 bp band was stronger and the 200 bp band was fainter.














When I saw this result I decided to look closer at my oligo design. After closer analysis, I looked that a possible misspriming event was possible.


A hypothesis that I had at the momment was that maybe the PCR oligos were too small (15 – 16 bases) and they were mispriming somewhere in the template. I designed new primers 22 – 25 bases long that were more specific to the end of the annealed/extended fragment. However, amplification with this primers should a strong 250 bp band, with  very faint bands at higher molecular weight (e.g. 300, 350 bp)


This misspriming event was already described in past posts. I sequenced the 250 bp fragment to see if I can discover any evidence of misspriming, however, traces showed not any evidence of missprimming.  What they did showed was that traces were not very good. peaks were very close to each other, with shoulders overlapping, and in some cases, there were small peaks in the bottom.


My next strategy was to clone the PCR amplified 250 bp band and the 200 bp band into a pgem vector. Then I could sequence a clone with each band size using primers that would bind in the vector. This strategy would allow me to sequence the ends of the fragment accurately.

The entire fragment of the clone from the 250 bp band was sequenced, and the traces were very clear. This fragment was in reality 234 bp long and it had a duplication of the last 34 bases from the fragment right end.





Now that I know exactly why my synthetic fragment showed extra 34 bases, I only need to understand why it happened and how can I fix it.

The first idea that comes to my mind is a polymerase slippage caused by some type of hairpin structure in the template or the copied DNA strand. But I have no idea exactly of how this could be happening. I can only think of something like the polymerase slippage model that happens in minisatellite/microsatellite evolution


Sorry I could get a better figure!

My next idea was to test is there are places within the fragment where hairpins could happen. So, I took the right oligo and I run a harpin analysis using the oligoanalyzer program from IDT.

I took the hairpin structure with highest free energy, -15.03 ΔG(kcal.mole-1),   and  I saw several places that could potentially form hairpins within the right oligo.




Three potential places where hairpin could occur are shown. Place 1 hairpin happens in the illumina sequencing priming site (purple). If this hairpin would be happening then I would be in big trouble since I cannot change that region without changing the design entirely. The second place, the longest, happen within the extra bases of the oligo (dark green). This hairpin is key because is very close to the place that was duplicated  (pos 38). If this hairpin was the cause of the extra bases, then the solution would be to redesign only one oligo. The last hairpin is located in the overlapping region with the other oligo. If this hairpin were responsible of the  extra 34 bases, less likely since it is farther away, the solution would be to redesign both oligos.

Finally, I can try is to use a PCR additive that will reduce secondary structures in the template DNA. Biosizebio website describes two additives that I could use: DMSO and glycerol. However, even if these additives work, I might have to redesign at least one oligo since the hairpin structures might interfere with further steps in which adapter and barcodes will be added to the fragments by low cycle PCR .

primNote:Primers with Illumina adapters (in yellow and blue) and barcodes (in grey and light blue) are shown annealed to the strands of the synthetic fragment