Keeping in mind your given probabilities for genome sequencing costs, how quickly will we learn what genes tend to make people into good researchers?

Phrased differently: how much more productive will individuals created using multi-generational in vitro embryo selection be when they grow up, given our knowledge at the time, and given that the technology is permitted and used? (If you think it won't be, that will be addressed in the following question.) Consider a productivity of 1 to be equivalent to that of the average (mean) non-embryo-selected scientist.

We acknowledge that children selected for traits contributing to research ability may be a minority of children selected for any reason, but assume that at least some group will still be selected for traits such as working memory, curiosity, and so on, and state how much more productive they'll be if they choose to go into research.

• Claim: Above a certain level of cognitive ability, further increases have minimal value. For instance, the famous California Terman study of high-IQ youth failed to include two future Nobel laureates in physics because they were somewhat below its IQ cutoff for entry.
  Implication: Selection for general cognitive function is unlikely to have great effects on scientific output.
  Sources: Gladwell, Malcolm. Outliers. New York: Little, Brown, and Company, 2008.

• Claim: The personality trait of openness to experience, which involves active imagination, aesthetic sensitivity, attentiveness to inner feelings, preference for variety, and intellectual curiosity, which makes important contributions to research ability, is known to be heritable, and therefore could be selected for in multi-generational in vitro embryo selection procedures. Identical and fraternal twins reared apart correlated .43 and .23, respectively, for Openness, while identical and fraternal twins reared together correlated .51 and .14. This was the highest correlation of any of the five major dimensions of personality.
  Implication: To the extent that psychological Openness correlates to research ability, it can be selected for fairly effectively.
  Source: Jang et al. "Heritability of the Big Five Personality Dimensions and Their Facts: A Twin Study". Journal of Personality 64:3, September 1996.
  <http://www.psych.umn.edu/courses/fall06/yoonh/psy3135/articles/Jang%20et%20al_1996.pdf>

• Claim: The heritability of IQ can be estimated at anywhere between 0.4 and 0.8, depending on the data used, e.g. twin studies, correlations between siblings, and adoption studies.
  Implication: IQ, which contributes to research ability, is at least somewhat heritable, so extensive embryo selection could increase it significantly. However, the impact of such an increase would depend on the degree to which science is limited by the supply of highly intelligent talent and how to estimate the effects of embryos with more favorable genetic predispositions to cognitive ability than any past humans have enjoyed.
  Source: Neisser et al. "Intelligence: Knowns and Unknowns". American Psychologist. Vol 51(2), Feb 1996, 77-101.
  <http://psychnet.apa.org/?fa=main.doiLanding&doi=10.1037/0003-066X.51.2.77>

• Claim: "The research, led by Robert Plomin of the Institute of Psychiatry in London, identified six genes that were strongly associated with high or low intelligence, but even the most powerful of these accounted for just 0.4 per cent of the variation in intelligence between individuals. The six together accounted for about 1 per cent of the variation in intelligence. Dozens of previous studies on twins and adopted children have established that about half of the variation in intelligence is down to environment." (2007)
  Implication: Though there are correlations between abilities useful for research and genetic variations, any individual genetic variation explains only a very small portion of differences between people in cognitive ability. Very large studies sequencing the genomes of hundreds of thousands or millions of individuals may be required to identify a substantial fraction of relevant variation. Some variants will be so rare that statistics cannot be assembled, forcing researchers to infer their effects from functional and other information, taking years after large-scale genome sequencing data become available.
  Sources: Coghlan, Andy. "'Intelligence genes' reveal their complexity." New Scientist 29 Nov. 2007. <http://www.newscientist.com/channel/being-human/mg19626324.100-intelligence-genes-reveal-their-complexity.html>.
  L M Butcher, O S P Davis, I W Craig, and R Plomin. "Genome-wide quantitative trait locus association scan of general cognitive ability using pooled DNA and 500k single nucleotide polymorphism microarrays." Genes, Brain, and Behavior 7 (2008): 435-446. <http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2408663>

• Claim: "Talent-search participants (286 males, 94 females) scoring in the top 0.01% on cognitive-ability measures were identified before age 13 and tracked over 20 years. Their creative, occupational, and life accomplishments are compared with those of graduate students (299 males, 287 females) enrolled in top-ranked U.S. mathematics, engineering, and physical science programs in 1992 and tracked over 10 years. By their mid-30s, the two groups achieved comparable and exceptional success (e.g., securing top tenure-track positions) and reported high and commensurate career and life satisfaction. [...] Individuals identified solely on the basis of one very high SAT score before the age of 13 achieved occupational success comparable to that of individuals attending world-class mathematics, science, and engineering graduate training programs."
  Implication: If there is a substantial correlation between genes and research ability, and this genetic material can be used to select embryos that will grow into adults with this exceptional ability, then the practical impact could be quite large. The ability multiplier for embryo-selected children could be as high as 100 or even 1,000.
  Source: Lubinski, David, Camilla P. Benbow, Rose Mary Webb, and April Bleske-Rechek. "Tracking Exceptional Human Capital Over Two Decades." Psychological Science 17 (2006): 194-99. Retrieved 9 Aug. 2008. < http://www.vanderbilt.edu/Peabody/SMPY/PsychScience2006.pdf>.

• Claim: "Within the United States, the regional differences in intellectual accomplishment have been nothing shy of astonishing...The population of Texas has exceeded that of New England for the last hundred years by a factor of 3 or 4...But a visit to Who's Who establishes that Texans have achieved virtually nothing in the sciences or philosophy (though there have been notable achievements in literature, music, and the arts in recent decades)."
  Implication: Research output critically depends on a culture of science: increasing abilities may not be channeled into advancing the state of scientific knowledge.
  Source: Nisbett, Richard. Intelligence and How to Get It. New York: Norton. 2009.