A general objective of these questions is to make you think precisely about the property of molecules (exactly what specific enzymes do, precise DNA sequences to be used or generated etc.) and numbers of molecules involved, as well as the diversity of outcomes (including unwanted outcomes) in a given procedure.

1. You are given some purified DNA, which you are told is a double-stranded circular plasmid. You are asked to cut it with three different enzymes (separately) and run the three sets of products along with DNA markers on an agarose gel. You do so, with equal aliquots and following the appropriate protocols carefully.

(a) In one lane (digest with EcoRI) you see one band only, lining up roughly with a marker of 8kb. You may be tempted to conclude that the plasmid is 8kb long. Describe TWO possible reasons why this conclusion may be incorrect AND, for each, what additional observations would likely reveal whether the alternative possibility you describe is correct or not.

(b) In another lane (digest with BamHI) you see a band of about 4kb and one of about 2kb. The two bands are of equal intensity. What do you conclude?

(c) You (or a machine) measure and plot the exact positions of DNA markers on a graph and read off from that graph the exact predicted size for the roughly 8, 4 and 2kb bands described above. Imagine one answer is 4,127bp. How accurate do you think that answer is AND will the accuracy likely be better for the 2, 4 or 8kb bands?

(d) In the third lane (digest with KpnI) you see bands of 4kb, 3kb and 0.8kb. What do you expect are the intensities of the 3kb and 0.8kb bands relative to the 4kb band? Explain.

2. Single-strands of DNA or RNA can hybridize to each other to form duplexes provided they have complementary sequences. This is basically because the sum of a number of favorable energetic contributions (H-bonds and base-stacking) is greater than negative ones (phosphodiester backbone repulsion and entropy). However, this is not an all or none situation. The equilibrium between single-strands and duplexes depends on the length of complementary sequence and GC vs AT content because G:C interactions are stronger (in addition to nucleic acid concentrations and chemical conditions). As a rough guide, for hybrids in the range of 14020bp each GC bp increases the melting temperature (50% dissociation) by 4C and each AT bp by 2C. Moreover, mis-matches and even single-stranded loops of various sizes will push the equilibrium away from duplexes but they are tolerated (surprisingly well in my opinion) in duplexes that have plenty of neighboring complementary sequences.
Sometimes it is advantageous to allow imperfect hybrids to form and other times it is important that perfect hybrids are highly favored. Stringency of hybridization may be adjusted accordingly (by varying temperature and/or salt concentration).

(a) You can adjust chemical conditions to maximize the stability of hybrids with 6 out of 6 bases pairing relative to 5 out of 6 (e.g. 5’ GATCTG 3’ binding better to 5’ CAGTC 3’ than to 5’ CGGTC 3’) and similarly for a variety of other lengths with diminishing success of single mis-match discrimination the longer the hybrid. However, even under the best conditions for every 100 correct hybrids there will probably be at least 5 imperfect hybrids. DNA ligase achieves far greater fidelity of mis-match discrimination with potentially complementary single-stranded regions as small as 1-4nt (ligating almost exclusively sequences that are exactly complementary) and DNA polymerase does the same (very few incorrect bases incorporated into DNA). How come?

(b) We discussed a strategy for assembling multiple chemically synthesized DNA oligonucleotides into much larger double-stranded DNAs by designing oligos with complementary sequences. Imagine all of your oligos are 100nt long and you want to create a molecule of 10kb with no gaps (the ends can have single-stranded regions). You consider three choices (please note that “hybridization” refers to a chemical process under ideal chosen conditions and “ligation” implies the presence of ligase carried out under conditions appropriate for the enzyme):

(i) One possibility is to hybridize pairs of oligos together to leave sticky ends (as for products of restriction enzyme digestion). You would then ligate these sticky fragments together in the order you want by having carefully designed which short (4nt or less) single-stranded overhangs are compatible. You would like to perform all of the ligations at the same time in one tube if possible to assemble the whole 10kb product.
What are TWO potential disadvantages of this method that limit its potential success or efficiency?

(ii) A second possibility is to hybridize all designed oligos together simultaneously. After that, add DNA ligase to seal nicks.
How would you choose to stagger the overlapping sequences? Would you just overlap each oligo pair by 50nt or would you take into consideration the DNA sequence of overlapping regions? Explain (and remember you task is to build a 10kb DNA of exactly prescribed sequence.

(iii) A third possibility is exactly as above except that you include DNA ligase during the hybridization (you might have to use Taq DNA ligase, which is thermostable and adjust the optimal temperature for hybridization to have salt concentrations compatible with ligase activity but you should assume that you can have optimal hybridization conditions and normal ligase activity at the same time). Is this likely to be better or worse than having ligation only after hybridization?

(iv) For any of these strategies to work do you need to do anything to the oligos after synthesis and purification? Explain.

(v) Each connecting step may be very efficient in the strategies above but it will never be 100%. So, what impurities do you expect to find AND how might you purify your correct product?

(vi) Would you expect your purified product (above) to all be exactly correct? Explain.

(vii) In principle how could you proceed further to generate a product that you would expect to be 100% correct (or very very close)?

3. This question is about standard Sanger dideoxy sequencing of DNA (we will consider other methods later). Sanger dideoxy cycle sequencing (sometimes called automated fluorescent sequencing) is routinely accomplished nowadays for individual sequencing runs by sending materials to a commercial service and receiving sequence files a day or two later (earlier versions of this methodology were introduced in class for historical purposes- slab gels, radioactive labeling etc. and are no longer used).

(a) Replacing radioactively labeled dATP with four ddNTPS, each labeled with a different fluorophore brought many benefits but it had the downside of lower sensitivity. Consider using fluorescently labeled dATP and unlabeled ddNTPs instead of four varieties of fluorescent ddNTPs in sequencing reactions.

(i) Would that increase sensitivity? Explain.

(ii) Which of the important benefits of moving from using radioactive dATP to fluorescent ddNTPs would be retained and which would be sacrificed? You should consider at least 3 benefits: the best answers will prioritize the most important benefits and explain each benefit precisely.

(iii) Some sequences in a template can cause local hairpins via intramolecular base-pairing (generally GC-rich sequences). Often DNA polymerase encountering such structures will pause and it will sometimes dissociate from the template and not re-initiate synthesis there. In such situations what would you expect to see in the DNA sequencing output if you were using fluorescent dATP AND if you were using standard fluorescent ddNTPs?

(b) Imagine you provide 400ng of a pure 4kb double-stranded DNA template for a standard sequencing reaction (normal four differently labeled ddNTPs) that involves 30 cycles. Everything works very well and you recover at least 800nt of good DNA sequence with most peaks of equal intensity. Use the approximation that one bp has a molecular mass, Mr of 700 (please DO use this number and not a more accurate one) and remember that the Mr of a molecule equals the mass of one mole (in grams) and that there are roughly 6 x 1023 molecules in one mole. Approximately how many DNA molecules are in a single peak of the DNA sequencing products? Explain your calculation line by line.

(c) Imagine, in the example above of sequencing a 4kb fragment, your detection techniques were hugely more sensitive so that you could reliably detect even just a single molecule in a “peak”. What do you think would be the minimum number of total molecules of DNA sequencing products that you would need to generate in order to produce a reliable DNA sequence?

(d) Returning to normal conditions of Sanger sequencing (i.e. four ddNTPs with different fluorophores and normal detection)…..
you will get good DNA sequence if you use the appropriate amounts of template and primer, if both are pure and if the primer only hybridizes significantly at only one position. For the purpose of these questions consider also that the DNA used as template will be double-stranded prior to denaturation (even though only one strand is depicted as “the template”) and that the thermostable DNA polymerases used do not have strong 3’ to 5’ exonuclease activity.

If your template and primer were as shown below, how would the returned DNA sequence read theoretically? Always indicate the 5’ to 3’ direction and begin at the beginning of what would be experimentally determined DNA sequence (even though the first few nucleotides of this sequence are not in fact read cleanly for minor technical reasons).

5’ GATGACTGCATTACTTATCA 3’
3’…AGATCTCTACTGACGTAATGAATAGTGCATGCGGACTAGCATTTGGACCT… 5’

(e) Consider some possible alterations to the primer shown above and explain what would be the consequences for the quality of your DNA sequencing results (try to give some idea of how significant any changes may be):
(i) deletion of the central ATT sequence (so the sequence is 5’ GATGACTGCACTTATCA)
(ii) deleting the 5’ G residue
(iii) changing the 3’ residue from A to T
(iv) deleting the 3’ A residue
(v) mixing the normal oligo with an equal amount of the oligo described in (ii)
(vi) mixing the normal oligo with an equal amount of the oligo described in (iv)
(vii) mixing the normal oligo with half as much of the complementary oligo (5’ TGTAAGTAATGCAGTCATC 3’)

(viii) If you requested an oligonucleotide of the correct sequence to use as a primer but did not request purification after synthesis which of the situations listed above would likely be most relevant (though the magnitude of the problem would be far lower)? Explain.

For some parts of this question you may wish to re-read the introduction to (d).

(f) The template will often be a plasmid DNA that you have made (because you are verifying its DNA sequence). Quick efficient plasmid DNA isolation methods are very good at removing most E.coli genomic DNA (“alkaline lysis method”) but imagine that you do a plasmid prep that has a relatively high contamination with E. coli genomic DNA (this could most likely happen when purifying a low copy number plasmid or a BAC). Will the presence of E. coli DNA likely affect the DNA sequence output? Explain (but remember to think carefully about the likely physical nature of the E. coli DNA- draw it to force yourself to do that).

(g) Imagine the sequencing output in an experiment is very clear but at one position you see equal-sized peaks of both C and T. What do you think is the most likely type of template that was used? Explain (exotic explanations will not receive full credit).

(h) An inexperienced researcher provided a sequencing template after checking a sample on a gel to see that there was a nice strong single band of PCR product of the expected length (i.e. the template looks good but you are not confident of whether the PCR product was “cleaned up” appropriately before sending off for sequencing). {PCR generally involves amplification of a segment of DNA from a small amount of starting DNA by using DNA polymerase and an excess of two primers that hybridize at two different places on the DNA segment to be amplified}. The sequencing output has two peaks instead of one at multiple positions (roughly 75% of all peaks). What likely happened?

A general objective of these questions is to make you think precise...