Directed evolution of polymerases that can read and write extremely long sequences

Supplemental Project Summary (derived from the original, changes underlined)Advances in synthetic biology have accelerated to the point where the synthesis of entire genomes is nowpossible. However, the technologies for these feats are painstaking, and the production of a new chromosomeor genome requires multiple years of effort, working from small fragments to ever larger assemblies. The speed(and ultimately scale) of large fragment assembly would be greatly improved if it were possible to routinelyamplify very long stretches of DNA (> 100 kb) in vitro. The methods developed in the execution of this proposalshould also prove extremely useful for greatly improved reagents for molecular diagnostics for SARS-CoV-2. Tothat end, this proposal is focused on the further development of a novel directed evolution method known asCompartmentalized Self-Replication (CSR), in which polymerases expressed in cells in emulsions undergothermal cycling to amplify their own genes, to generate long read DNA polymerases that should prove capableof generating PCR amplicons > 100 kb in length, with few errors. To achieve this goal, we propose to develop anovel library construction method that most efficiently brings together sequence and structural domains from avariety of DNA polymerase variants to form diverse chimeras (Aim 1.1), and to sieve these libraries usingimprovements to CSR that will allow us to select for extreme processivity in yeast (Aim 1.2) and efficient error-correction (Aim 1.3). Using the methods in Aim 1.2, we can produce polymerase variants that should be able todirectly participate in RT-qPCR without sample preparation, including from samples inactivated with denaturants.The variants that result will be characterized for their ability to synthesize long amplicons in vitro (Aim 2.1), fortheir fidelity (Aim 2.2), and for their detailed kinetic properties (Aim 2.3). Finally, to better ensure the processivityof the resultant polymerase chimeras, we will append either DNA-binding domains (Aim 3.1) or clamps (Aim3.2) that should lead to much better ability to grip DNA. Using the methods described in Aim 3.1, we can generatethermostable reverse transcriptases that should prove useful for the development of isothermal amplificationassays that can be used at point-of-care, or in resource-poor settings. In addition to accelerating the ongoingrevolution in genome synthesis, such long-read polymerases should also pave the way to new sequencingtechnologies, including for single molecule sequencing and for single cell sequencing. In the current crisis,polymerase engineering for particular functions, directed towards needs that the community has and that needto be resolved for forward motion on testing, is a critical component of a national plan.