Relative Generality and Risk: Quantitative Measures for Broad Catalyst Success

The
performance of chiral catalysts is typically evaluated
against
empirical reaction outputs like yield and selectivity with traditional
analyses limited to a single model system. Expansion of the reaction
space permits catalysts to be assessed for generality, and this provides
another useful metric for measuring the effectiveness of a catalyst.
The catalyst generality algorithm will assign quantitative generality
values to catalyst structures, but such broad assessments are applied
with the assumption that the reactions under evaluation are more or
less the same by disregarding any inherent challenges associated with
a particular reaction class. To address this limitation, we introduce
two metrics: relative generality and risk. These are designed to correct
for variations in reaction difficulty and enable a more nuanced evaluation
of the catalyst performance relative to the specific demands of each
reaction. We show in a number of challenging examples that these metrics
allow researchers to distinguish between catalysts genuinely exhibiting
superior performance and those appearing favorable due to application
toward less demanding reactions. This represents a significant advancement
in quantifying catalyst success with demonstrated applications in
retrospective analyses and early insights into emerging catalyst classes.