Chemistry informer libraries: A chemoinformatics enabled approach to evaluate and advance synthetic methods

Article abstract of DOI:10.1039/c5sc04751j

Major new advances in synthetic chemistry methods are typically reported using simple, non-standardized reaction substrates, and reaction failures are rarely documented. This makes the evaluation and choice of a synthetic method difficult. We report a standardized complex molecule diagnostic approach using collections of relevant drug-like molecules which we call chemistry informer libraries. With this approach, all chemistry results, successes and failures, can be documented to compare and evolve synthetic methods. To aid in the visualization of chemistry results in drug-like physicochemical space we have used an informatics methodology termed principal component analysis. We have validated this method using palladium- and copper-catalyzed reactions, including Suzuki-Miyaura, cyanation and Buchwald-Hartwig amination.

Full text of DOI:10.1039/c5sc04751j

Chemical
Science
View Article Online
View Journal
EDGE ARTICLE
Chemistry informer libraries: a chemoinformatics
enabled approach to evaluate and advance
synthetic methods†
Cite this: DOI: 10.1039/c5sc04751j
Peter S. Kutchukian,a James F. Dropinski,b Kevin D. Dykstra,c Bing Li,c
Daniel A. DiRocco,b Eric C. Streckfuss,c Louis-Charles Campeau,b Tim Cernak,c
c
b
c
b
*
*
Petr Vachal, Ian W. Davies, Shane W. Krska and Spencer D. Dreher
Major new advances in synthetic chemistry methods are typically reported using simple, non-standardized
reaction substrates, and reaction failures are rarely documented. This makes the evaluation and choice of
a synthetic method difficult. We report a standardized complex molecule diagnostic approach using
collections of relevant drug-like molecules which we call chemistry informer libraries. With this
approach, all chemistry results, successes and failures, can be documented to compare and evolve
synthetic methods. To aid in the visualization of chemistry results in drug-like physicochemical space we
have used an informatics methodology termed principal component analysis. We have validated this
method using palladium- and copper-catalyzed reactions, including Suzuki–Miyaura, cyanation and
Buchwald–Hartwig amination.
Received 9th December 2015
Accepted 15th January 2016
DOI: 10.1039/c5sc04751j
www.rsc.org/chemicalscience
different synthetic methods. In this work, we have created two
substrate collections that are relevant to pharmaceutical
Introduction
research, which we have termed chemistry informer libraries, to
evaluate this standardized diagnostic approach.2 Using these
libraries, consisting of medium-complexity aryl pinacol boro-
nates and high-complexity aryl halides, we have conducted
a systematic study comparing the relative success rates of
various reported catalytic methods in order to identify the best
rst-pass options for synthesis of pharmaceutically relevant
compounds. We also demonstrate how these informer libraries
can serve as diagnostic tools to improve the performance of
chemistry methods in order to achieve broader utility and
discuss how this approach may be combined with chemo-
informatics tools to enable in depth explorations of structure–
reactivity space.
Typical synthetic publications employ simple model
substrates to test the effects of local steric and electronic vari-
ations (Fig. 1). Recognizing the limitations of this approach,
especially in regards to functional group compatibility, Glorius
recently developed a diagnostic method in which molecular
fragments are added to simple substrate combinations to
determine which functionalities interfere with chemistry
performance.3 This tool is operationally easy to execute and
gives ready access to valuable reactivity data; however, we have
found that approximating the chemical reactivity of a complex
molecule as that of a collection of disconnected fragment
structures oen fails to account for important aggregate
molecular properties, such as chelation effects and substrate
solubility, that only arise in the context of whole molecules. In
contrast, the chemistry informer library evaluates not only local
Organic synthesis continues to advance at a rapid pace, giving
rise to new synthetic methods that enable transformations that
would have been unthinkable only a few years ago. These new
methods hold promise of applications across a diverse range of
endeavours, from natural product synthesis to pharmaceuticals
and materials, provided they work on the types of complex
chemical structures commonly encountered within those
disciplines.1 Most reported synthetic methods use simple
commercial model substrates that perform in high yield to
demonstrate scope, and there is little overlap between substrate
sets from one publication to the next, making comparisons
between methods difficult. For the chemist needing to choose
among various reported methods to accomplish a given trans-
formation in a complex synthesis, this incomplete and frag-
mented data set represents a signicant barrier to utilization.
To address this problem, we envisioned assembling collections
of standard substrates representative of structures encountered
in complex synthesis and using these to document the complete
chemistry performance (both successes and failures) of
aDepartment of Structural Chemistry, Merck Research Laboratories, Merck and Co.,
Inc., Boston, MA 02115, USA
bDepartment of Process and Analytical Chemistry, Merck Research Laboratories, Merck
and Co., Inc., Rahway, NJ 07065, USA. E-mail: spencer_dreher@merck.com
cDepartment of Discovery Chemistry, Merck Research Laboratories, Merck and Co.,
Inc., Rahway, NJ 07065, USA. E-mail: shane_krska@merck.com
† Electronic supplementary information (ESI) available. See DOI:
10.1039/c5sc04751j
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
atoms and rotatable bonds compared to the literature
compounds. We highlight the seminal Suzuki–Miyaura
coupling paper8 (Fig. 2A, red) to demonstrate that more recent
Suzuki–Miyaura coupling methodology papers,9 as well as the
leading Buchwald–Hartwig C–N coupling publications and the
more recently reported aryl boronate cyanation reaction (all in
blue), occupy a broader coverage of pharmaceutically relevant
chemical space. Seen in this context, our goal in creating the
chemistry informer libraries was to assemble compound test
sets that could claim an even higher complexity and broader
coverage of drug-likeness as demonstrated using PCA. Beyond
this, we aspired to use chemoinformatics tools such as PCA to
gain deeper insight into the reactivity patterns that would result
from utilization of informer libraries with different synthetic
methods, and use this knowledge to improve the applicability of
modern synthetic methods to drug synthesis.
Fig. 1 Synthetic chemistry diagnostic approaches. Standard literature
reports typically demonstrate simple, high-performing substrates; the
Glorius “robustness” test probes the effects of potentially interfering
molecular fragments on chemistry performance, and in this work we
examine how standardized collections of full, complex molecules can
be used to compare and improve literature synthetic methods.
Aryl and heteroaryl pinacol boronate informer library
To explore the merits of the chemistry informer library concept,
we curated a collection of heterocycle-containing pinacol bor-
onates that contained the most commonly encountered rings in
drugs10 as well as functional groups that are important in drug
structures, such as amines, amides, sulfonamides, sulfones,
alcohols, nitriles and aryl halides (Fig. 3A). Thousands of
diverse pinacol boronates are available from commercial
vendors, largely driven by their use in the venerable Suzuki–
Miyaura reaction, popular in medicinal chemistry labs for
decades.11 The Suzuki–Miyaura coupling is well established to
work well on pharmaceutically relevant structures, so we chose
this reaction to validate the performance of the boronate
informer library and set a benchmark to which other methods
could be compared. Since the discovery of the Suzuki–Miyaura
reaction 35 years ago, signicant efforts by many laboratories
have led to new generations of Pd catalysts.12 We chose to
examine whether the evolution of the Suzuki–Miyaura coupling
methodology could be correlated, using our boronate informer
library, with increased applicability in the pharmaceutical
space. To this end, using the retrospective literature analysis as
a comparative backdrop (Fig. 3B, red and blue dots), we created
a standardized test set of potential Suzuki–Miyaura products
with the boronate informer library, using 6-bromoindole as the
coupling partner (Fig. 3B, yellow dots), for which the PCA
indicates overlap of chemical space with the most evolved
literature as well as increased coverage of relevant drug-like
space that was not systematically explored in synthetic methods
papers.13
steric and electronic effects and functional group compatibility,
but also complex, whole molecule interactions.
Results and discussion
Principal component analysis
In order to assemble chemistry informer libraries representa-
tive of problems encountered in pharmaceutical research, we
required a means to assess these molecules and their potential
reaction products in the context of the chemical space occupied
by actual small molecule drugs. To this end, we employed
principal component analysis (PCA), an informatics method
that has proven fruitful in reducing complex multidimensional
chemical space to two or more dimensions that capture the
variation in the higher dimensional space, allowing meaningful
visualization.4 Performing a PCA on physicochemical properties
has proven especially useful in evaluating relationships
between sets of compounds.5 We trained our PCA with 14
physical chemical properties including molecular weight,
number of rings, hydrogen bond donors, hydrogen bond
acceptors, fraction sp3 character and Alog P.6 We rst used this
PCA to compare the chemical space of FDA-approved small
molecule drugs with products from the literature reports for the
three different synthetic methodologies that are explored in this
paper. We visualize the rst two principal components (Fig. 2A,
58% of variance explained), and also depict the physical prop-
erty trends that represent the greatest loadings for PC1 (Fig. 2B
Suzuki–Miyaura reaction
and C) and PC2 (Fig. 2D and E) for this space.7 This analysis With this test set, we evaluated several landmark Suzuki–
reveals that literature compounds are generally clustered tightly Miyaura reaction methods from the literature to systematically
in one area of chemical space due to their low molecular weight, compare their performance (Fig. 3D). In order to facilitate these
and for their given size, high aromatic content, high lip- studies, experiments were conducted using parallel microscale
ophilicity, and low number of hydrogen bond donors/acceptors reaction techniques developed previously in our laboratories.14
relative to marketed drugs. Furthermore, from analysing addi- These experiments demonstrated that even the earliest pub-
tional components (PC4 vs. PC3, ESI†), we see that some drugs lished method, using tetrakis(triphenylphosphine) palladium
have greater variation in properties, such as number of stereo (5 mol%) as the catalyst with inorganic base in an ethereal
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
Fig. 2 Principal component analysis (PCA) comparing small molecule drugs with products described in synthetic literature reports. (A) shows
a principal component analysis used to visualize the physicochemical space occupied by marketed small molecule drugs compounds (grey) and
products described in literature reports for the several reaction types evaluated in this work: in red are compounds that were prepared in the
seminal Suzuki–Miyaura paper, and in blue are the products formed in leading literature Suzuki–Miyaura reports, recent methods reported for
the conversion of aryl pinacol boronates into aryl nitriles and recent Buchwald–Hartwig C–N coupling methods. Representative structures from
both the small molecule drugs collection and literature reports are highlighted. Fig. 2B–E depict how properties with the greatest contribution to
PC1 and PC2 are mapped by the PCA, with black representing the highest values for each parameter. See ESI† for full details.
solvent at elevated temperature, gave synthetically useful15 At room temperature this highly active catalyst provided only 18
amounts of products with all but two of the informer library products in useful yield, although increasing the temperature to
members when 6-bromoindole was used as the coupling 100 ꢀC gave success with 2 additional problematic substrates.18
partner, with an average yield across the library of 46% (Fig. 3D, Taken as a whole, the results of the informer library study of the
entry 1). Subsequently, we tested (1,10-bis(diphenylphosphino) Suzuki–Miyaura reaction validated the performance of the
ferrocene)palladium(II)chloride [(dppf)PdCl2] under similar individual library members and conrmed the generality of the
conditions and found that this bidentate phosphine palladium Suzuki–Miyaura methodology for pharmaceutically relevant
catalyst that has found extensive application in natural product substrates. These studies also showed that, despite the impor-
and drug synthesis16 gave successful outcomes with all 24 tant extension to aryl chlorides enabled by ligands such as
members and a higher overall average yield (entry 2).17 Despite XPhos, certain substrates, including oxadiazole B9 and tetrazole
their strengths, both of these catalysts exhibit limited scope B12, appeared to perform better with older generation catalysts
with aryl chloride substrates. Buchwald and co-workers recently in conjunction with aryl bromide coupling partners.
reported a highly reactive system (XPhos G2 precatalyst, THF,
aqueous K3PO4) that is capable of coupling sensitive boronic
Pinacol boronate cyanation reaction
acids with aryl halides, including chlorides, at room temper-
We next sought to identify a relatively new transformation that
ature.9e Since it was not clear at the outset whether room
has received much less “eld testing” on molecules of signi-
cance to pharmaceutical synthesis. Recently, the discovery of
direct C–H borylation methods19 has improved access to pinacol
temperature would be optimal for boronate ester coupling
partners with such diverse structures, we ran the informer
library, now with 6-Cl-indole as the halide coupling partner, at
boronates, enabling their use as part of late stage functionali-
room temperature, 60 ꢀC and 100 ꢀC (Fig. 3D, entries 3, 4 and 5).
zation strategies wherein the boronate functionality becomes
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 3 Aryl and heteroaryl pinacol boronate informer library. (A) displays the 24-member boronate ester library plated in 1 mL vials in 10 mmol
quantities, used for Suzuki–Miyaura and cyanation chemistry evaluation. (B) and (C) display the principal component analysis of potential
products formed from Suzuki-Miyaura coupling and cyanation reactions, respectively, of the informer library (in yellow), literature products (in
blue and red), and marketed drugs (in grey). (D) gives experimental results for Suzuki–Miyaura coupling reactions (entries 1–5) and cyanation
reactions (entries 6–12) of the informer library members along with statistics for each method. For experimental details, see the ESI.†
a gateway to introduce a diverse array of groups, such as uo- reactions in the literature (Fig. 3C) revealed that the types of
rine, alkyl, peruoroalkyl, amines, ethers, etc., that can modu- product structures used to exemplify the chemistry (shown in
late physicochemical and pharmacological properties.20 In blue) were for the most part low molecular weight, simple are-
conjunction with this interest in C–H borylation, several groups nes bearing little resemblance to drug-like molecules (shown in
have recently reported methods for the cyanation of aryl boro- grey). In contrast, potential products derived from the cyanation
nates to effect the late stage introduction of nitriles.21 Evalu- of pinacol boronate informer library substrates (shown in
ating the chemical space explored by boronate cyanation
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
yellow), although small in size relative to some drugs, sampled reaction employing piperidine as the nucleophile. Fig. 4B plots
a broader cross-section of drug-like structural chemical space. the PCA of reaction products in six representative C–N
We chose three recently published cyanation protocols with coupling methodology literature reports22 (shown in blue) as
diverse sets of reagents and conditions (Fig. 3D, entries 6, 7 and compared to drug-like chemical space (grey). In contrast, the
8) to evaluate against the boronate informer library. The potential reaction products resulting from C–N coupling of the
method published by Hartwig21a (using Zn(CN)2, Cu(NO3)2 and high complexity aryl halide informer library with piperidine
CsF base with a methanol/water solvent mixture at 100 ꢀC) gave (in yellow) are clearly more dispersed within pharmaceutically
the best results, with 13 products obtained in useful yields and relevant space that has not been systematically explored in
a 26% average yield across the library. The best literature previous reports.
performance still le much to be desired, however, especially
when benchmarked against the previous Suzuki–Miyaura
coupling results. We examined whether the informer library
Pd C–N coupling
diagnostic might also be applied to optimize this method to Over the past two decades C–N coupling of aryl (pseudo)halides
increase its scope. Decreasing the reaction temperature to 70 ꢀC has become one of the most widely used classes of reactions in
(entry 9) gave very similar overall results, while performing the pharmaceutical synthesis.11b,23 Generations of new catalyst
reactions at 40 ꢀC resulted in lower conversions across the development have resulted in highly evolved systems that are
library (entry 10). Our observation that the MeOH/water solvent capable of coupling a wide array of aryl halides and pseudo-
system was not ideal for dissolving the medium-complexity halides with diverse nitrogen-based nucleophiles such as
informer substrates led us to examine the use of solubilizing co- amines, amides, and N-heterocycles. Despite this progress, in
solvents. We found that running the reactions in the presence of our laboratories the C–N coupling reaction has exhibited
dimethylformamide (DMF) co-solvent resulted in signicant a relatively low success rate (ꢁ45%) when applied in the nal
increases in reaction yield and hit rate across the library. Using steps of drug synthesis. Using the high-complexity aryl halides
DMF at 70 ꢀC (entry 11) gave 15 out of 24 products and an informer library, we surveyed the landmark Pd C–N coupling
improved overall average yield of 36% across the library; literature, starting with the rst conditions reported by both
ꢀ
reducing the reaction temperature to 40 C (entry 12) gave the Buchwald and Hartwig employing ((o-Tol)3P)2PdCl2, NaOtBu
highest hit rate (17 out of 24 substrates) and a similar overall and base with toluene as the solvent at 100 ꢀC.22a,b These
average yield. These results support the use of informer libraries conditions gave a very low success rate against the informer
as a tool to evolve existing methods toward increased generality. library, resulting in formation of only 2 out of 18 products in
In this particular case, the use of drug fragment-like boronates greater than 20% yield (Fig. 4D, entry 1). Two subsequently re-
identied solubility as being a key factor in determining reac- ported Pd catalysts bearing electron-rich ligands capable of
tion outcome. Beyond this, the reactivity trends embedded activating aryl chlorides, one a Josiphos bis(phosphine) ligand
within the informer library results revealed some curious (entry 2)24 and the other a saturated carbene ligand (entry 3),22d
observations that could prove useful to researchers looking to showed only marginal improvements under similarly harsh
develop improved methods. For example, with the exception of reaction conditions. The electron-rich ligand RuPhos has been
tetrazole-containing boronate B12, which also gave poor results recommended as an optimal ligand for 2ꢀ amine coupling,25
in the Suzuki–Miyaura coupling, all of the poorest performing and advances in Pd precatalyst design have allowed for the use
members of the informer library (B1, B4, B13, B20 and B22) of milder bases and reaction temperatures with this ligand.26
contained the Bpin moiety in a meta position to a nitrogen atom The combination of RuPHOS G2 precatalyst and LiHMDS/
in a six-membered heterocyclic ring. Although the reasons for dioxane at 80 ꢀC resulted in 5 compounds being prepared (entry
this are not clear, future work on cyanation methods should 4). Varying solvent/base combinations (entries 5 and 6) with this
include examples of these types of structures given their prev- catalyst gave only marginal improvements. We recently reported
alence in drugs.
that the use of tBuXPhos G3 precatalyst in conjunction with the
strong soluble phosphazene base P2Et in the polar solvent
DMSO at ambient temperature gave broad scope for C–X
coupling reactions on complex substrates, including some of
High complexity aryl halide informer library
The results with the pinacol boronate informer library vali- the members of this informer library.27 This system exhibited
dated the concept of using collections of moderately complex low reactivity at room temperature (entry 7) but gave somewhat
drug fragments to systematically probe reaction utility. In better results at 60 ꢀC (entry 8), essentially equivalent to the best
order to explore the use of an informer library concept within RuPhos results. The results of this study demonstrated how
signicantly more complex drug-like space, we assembled advances in Pd-catalyzed Buchwald–Hartwig coupling protocols
a collection of 18 higher complexity aryl halides (3 chlorides, over time have resulted in incremental improvements in
13 bromides and 2 iodides) from the Merck & Co., Inc. applicability to drug-like intermediates. Since no method was
compound collection (Fig. 4A). These compounds represent able to give better than a 33% success rate, and even a hypo-
the penultimate intermediates from various drug discovery thetical parallel evaluation of all 8 sets of Pd-catalyzed C–N
and development programs and thus ideally typify pharma- coupling conditions would have resulted in only 50% of the
ceutical substrates. In order to demonstrate the utility of this compounds being prepared, this reaction is still far from
informer library, we chose to evaluate the C–N coupling a solved problem in high-complexity drug synthesis.
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 4 High-complexity aryl halide informer library. (A) displays the 18-member aryl halide array, plated in 2.5 mmol quantities in 250 mL
“microvials”, used for Pd and Cu C–N coupling chemistry evaluation. (B) displays the principal component analysis of potential products formed
from C–N coupling reactions of the informer library (in yellow), literature products (in blue and red), and marketed drugs (in grey). (C) shows
structures for the catalysts employed in the Pd and Cu C–N reactivity study depicted in (D). For experimental details, see the ESI.†
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
we found that fragments containing oxadiazole and 4-chlor-
ophenylsulfonamide functionalities signicantly diminished the
reaction yield of a model Suzuki–Miyaura coupling reaction,
mirroring the performance observed in the informer library. In
contrast, a tetrazole-containing additive did not inhibit the test
Suzuki–Miyaura coupling reaction, nor was it consumed in the
reaction. Similarly, we studied the failure of the C–N coupling of
halide X7 with piperidine, rst using the simplest fragment
approach, and found that none of the potentially interfering
fragments that could be derived from the deconstruction of
complex molecule X7 diminished the reactivity of a simple C–N
coupling. Instead, we considered it more likely that the local
steric and electronic demands of the core of X7 were responsible
for the low reactivity and conrmed that 1-benzyl-7-bromoindole
indeed showed very low (<1% LC area percent) product forma-
tion. These studies show that the simplest, most experimentally
accessible fragment-based method may signicantly under-
predict reaction failures in complex systems. It is likely that in
such cases the aggregate of steric and electronic properties of the
reacting core, the presence of various functional groups and the
overall physical properties of the entire molecule will prove
deterministic of reaction success.
Cu C–N coupling
Having explored Pd-catalyzed C–N coupling methods, we next
turned our attention to conditions employing Cu catalysts.
Although traditionally Cu-based methods have been employed
more commonly in the coupling of amide and N-heterocyclic
substrates, some literature precedent points to their utility in
amine couplings.23 Reported Cu-based systems possess several
attractive features when viewed from
a pharmaceutical
perspective, including the common use of polar aprotic solvents
and mild inorganic bases. We rst evaluated a variation of the
original Ullmann conditions28 which are stoichiometric in Cu
(Fig. 4D, entry 9): these conditions gave no hits across the
library. Moving on to one of the earliest reported examples from
Buchwald and coworkers22c of a Cu-catalyzed C–N coupling,
using CuI in conjunction with the diketone ligand L1 as the
catalyst in DMF solvent with Cs2CO3 base, gave two hits (entry
10). The Ma group has reported oxamate-based ligands that give
greatly increased reactivity in Cu-catalyzed amine couplings.22f
Using the reported conditions employing L2 as the ligand (entry
11) gave the best overall performance observed thus far, with 9
total products generated in usable yields. Hoping to improve on
these results, we prepared a series of analogs of the original
oxamate ligand L2 and used the informer library as a diagnostic
to determine whether any of these exhibited increased reactivity
and scope (entries 12–18). Ligand L4 gave similar results to L2,
with 8 compounds prepared in usable yields (entry 13). Noting
the extremely clean proles of these reactions, we evaluated
both the L2 and L4 systems again at 100 ꢀC, and found that L4
gave 9 compounds at 33% average yield, the best reaction
prole for any Pd or Cu conditions we evaluated (entry 17).
Further increases in temperature were not benecial (entry 18).
The successful coupling of 9 out of 15 aryl bromides and iodides
in the informer library with one set of conditions clearly
substantiates the notion that Cu is an excellent choice for C–N
coupling in highly complex synthesis, although signicant
room for improvement, including the extension to aryl chlo-
rides, remains.
The future of chemistry informer libraries
In this work we have used heat maps, shown in Fig. 3 and 4, to
provide a visualization of individual reaction performance as
well as reactivity patterns and trends for different synthetic
methods. As useful as such depictions may be, they ignore the
richness of structural information embedded in the individual
informer library members and products. Fig. 5 depicts one way
in which both reactivity and structural information may be
visualized together. By mapping the experimental yields
(magnitude represented by size of circles) for particular
informer library products onto the two-dimensional PCA of
marketed drugs, different chemistry methods and even entirely
different reaction types can be represented in a single graph to
reveal chemical reactivity trends and relevance within a given
region of structural space where the products share similar
physicochemical characteristics. Fig. 5 illustrates the possibili-
ties of systematically exploring structure–reactivity relation-
Comparison of the whole molecule informer library and the
fragment-based robustness test
The informer library approach to reaction appraisal carries high ships within complex chemical space with the aid of modern
experimental demands, requiring access to complex molecules chemoinformatics techniques. Obviously, the data sets
for experimentation as well as the need for scale-up, isolation and acquired in this work are much too limited to enable the
characterization of all new products in order to enable quanti- construction of quantitative, predictive structure–reactivity
tative LC analysis.29 By comparison, the simpler fragment-based models. Currently, despite signicant advances made in the
“robustness” test involves adding potentially interfering frag- high-throughput screening of reactions on a small scale30 the
ments to simple model reactions and then measuring the reac- number of chemistry informer library compounds that can be
tion performance relative to a control that does not have the evaluated is limited by practical considerations arising from the
fragments present. This method measures both the decrease in need to isolate and characterize products in order to quantify
product formation, indicating reaction inhibition, and the reaction yields. However, this limitation may not remain for
decrease in the amount of the fragment itself remaining at the long.31 In this case, one could then assemble and test the
end of reaction, indicating decomposition pathways. In order to reactivity of larger informer sets that would provide a statistical
compare the two diagnostic methods, we chose 3 boronates (B9, representation of diverse structural motifs in a complex mole-
B11 and B12) that failed in the Suzuki–Miyaura reaction with 6- cule setting. Mining the rich data sets that would result from
chloroindole using the XPhos G2 precatalyst system at 100 ꢀC (see such an exercise would provide a holistic picture of structure–
ESI† for details on all fragment-based experiments). In this study, reactivity relationships that would encompass the various levels
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
Fig. 5 Informer library reactivity data mapped onto relevant physicochemical PCA to produce a representation of structure–reactivity rela-
tionships. Each colored ball represents one molecule successfully prepared, and the size of the ball is proportional to the yield of the reaction. In
this way, different reaction types can be directly compared for performance and relevance within a region of structural space.
of detail found in Fig. 1, including: (1) information on local relationship between reaction types and diverse complex
steric and electronic effects at reaction centers, (2) identica- substrates. The general application of this approach would
tion of interfering structural elements, and (3) the effects of enable synthetic practitioners to make better decisions about
aggregate molecular properties, such as potential chelation which synthetic methods to prioritize in their problem solving
motifs or bulk physicochemical characteristics.
and give them a diagnostic that could reveal if a new method
was signicantly better than existing methods. Chemistry
informer libraries could also provide insight to academic
researchers, as well as funding agencies, into which synthetic
methods are mature versus those, such as the C–N coupling
reaction, that apparently still require signicant innovative
development to become broadly relevant to complex synthesis.
The same analyses can point out specic gaps in substrate
scope that need to be addressed. Finally, the chemistry informer
library concept points toward a new way to study chemical
reactivity, in which chemoinformatics tools are applied to large
Conclusions
We have demonstrated how a standardized chemistry informer
library can allow direct head-to-head comparisons of different
synthetic methods in complex structural space. In contrast to
current methodology studies that tend to explore the reactivity
of simple substrates, or total synthesis papers that tend to
report only the positive results for the reactivity of specic
complex substrates, our method comprehensively explores the
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016

View Article Online
Edge Article
Chemical Science
data sets generated from complex substrates to enable a holistic 10 R. D. Taylor, M. MacCoss and A. D. G. Lawson, J. Med. Chem.,
understanding of structure–reactivity relationships, from local 2014, 57, 5845.
steric and electronic effects at reaction centers to aggregate 11 (a) M. L. Crawley and B. M. Trost, Applications of transition
physicochemical properties arising from the entire substrate
structure. Efforts in our labs towards further validating this
chemoinformatics-enabled approach to reaction evaluation and
optimization are underway. We eagerly anticipate that other
metal catalysis in drug discovery and development: an
industrial perspective, Wiley, Hoboken, 2012; (b)
¨
D. G. Brown and J. Bostrom, J. Med. Chem., 2015, 58, DOI:
10.1021/acs.jmedchem.5b01409.
laboratories will begin to adopt and expand upon these 12 For reviews on the Suzuki–Miyaura coupling, see: (a)
methods, and in doing so help drive the eld to develop
synthetic transformations with ever more powerful applications
in complex molecule synthesis.32
N. Miyaura and A. Suzuki, Chem. Rev., 1995, 95, 2457; (b)
A. Suzuki, J. Organomet. Chem., 1999, 576, 147; (c)
F. Bellina, A. Carpita and R. Rossi, Synthesis, 2004, 2419;
(d) C. Valente and M. G. Organ, in Boronic Acids:
Preparation and Applications in Organic Synthesis, Medicine
and Materials, ed. D. G. Hall, Wiley, Chichester, 2nd edn,
2011.
Acknowledgements
We thank Meir Glick and Eric Fischer for helpful discussions,
Mikhail Reibarkh for assistance with NMR experiments and 13 While many Suzuki–Miyaura reactions have been reported in
Natalya Pissarnitski for assistance in compound purication.
the literature for complex drug-like structures or natural
products, there is no standardizedreactivity data for these
compounds, and in many cases no attempt is made to
determine the actual yields of reactions.
Notes and references
1 A. Nadin, C. Hattotuwagama and I. Churcher, Angew. Chem., 14 (a) C. S. Shultz and S. W. Krska, Acc. Chem. Res., 2007, 40,
Int. Ed., 2012, 51, 1114.
2 For related approaches, see: (a) K. Zhang, S. El Damaty and
1320; (b) S. D. Dreher, P. G. Dormer, D. L. Sandrock and
G. A. Molander, J. Am. Chem. Soc., 2008, 130, 9257.
R. Fasan, J. Am. Chem. Soc., 2011, 133, 3242; (b) E. N. Bess, 15 For a typical synthesis of an investigational drug compound
A. J. Bischoff and M. S. Sigman, PNAS, 2014, 111, 14698.
3 (a) K. D. Collins and F. Glorius, Nat. Chem., 2013, 5, 597; (b)
K. D. Collins and F. Glorius, Acc. Chem. Res., 2015, 48, 619; (c)
in a medicinal chemistry project, a solution yield of >20%
can reliably provide a usable amount of material for initial
biochemical testing aer purication.
K. D. Collins and F. Glorius, Tetrahedron, 2013, 69, 7817; (d) 16 R. Rossi, F. Bellina, M. Lessi, C. Manzini, G. Marianetti and
K. D. Collins, A. Ruhling, F. Lied and F. Glorius, Chem. Eur. J., L. A. Perego, Curr. Org. Chem., 2015, 19, 1302.
2014, 20, 3800; (e) K. D. Collins and F. Glorius, Nat. Protoc., 17 Selected Suzuki–Miyaura coupling conditions were repeated
2014, 9, 1348.
on a larger scale to conrm product structures and to obtain
UV response factors for solution yield determination. See
ESI† for more details.
4 J. L. Medina-Franco, K. Martinez-Mayorga, M. A. Giulianotti,
R. A. Houghten and C. Pinilla, Curr. Comput.–Aided Drug
Des., 2008, 4, 322.
5 A. A. Shelat and R. K. Guy, Curr. Opin. Chem. Biol., 2007, 11,
244.
18 The XPhos G2 precatalyst has been reported to give good
results in Suzuki–Miyaura couplings of arylboronic acids
with nitrogen-rich, unprotected heterocycles at elevated
temperature: see ref. 9g.
6 In order to generate a PCA visualization that was consistent
across each analysis, we performed a PCA on all compounds, 19 (a) I. A. I. Mkhalid, J. H. Barnard, T. B. Marder, J. M. Murphy
although in many of our gures only a subset of compounds
is depicted.
7 We visualize PC3 and PC4 in Fig. S1 in the ESI† (78%
cumulative variance explained).
8 N. Miyaura, T. Yanagi and A. Suzuki, Synth. Commun., 1981,
11, 513.
and J. F. Hartwig, Chem. Rev., 2010, 110, 890; (b)
J. F. Hartwig, Chem. Soc. Rev., 2011, 40, 1992; (c)
J. F. Hartwig, Acc. Chem. Res., 2012, 45, 864; (d)
S. M. Preshlock, B. Ghaffari, P. E. Maligres, S. W. Krska,
R. E. Maleczka and M. R. Smith, J. Am. Chem. Soc., 2013,
135, 7572.
9 (a) A. Zapf, R. Jackstell, F. Rataboul, T. Riermeier, 20 (a) M. A. Larsen and J. F. Hartwig, J. Am. Chem. Soc., 2014,
A. Monsees, C. Fuhrmann, N. Shaikh, U. Dingerdissen and
M. Beller, Chem. Commun., 2004, 38; (b) N. Kudo,
M. Perseghini and G. C. Fu, Angew. Chem., Int. Ed., 2006,
136, 4287; (b) T. Cernak, K. D. Dykstra, S. Tyagarajan,
P. Vachal and S. W. Krska, Chem. Soc. Rev., DOI: 10.1039/
c5cs00628g.
45, 1282; (c) C. J. O'Brien, E. A. B. Kantchev, C. Valente, 21 (a) C. W. Liskey, X. Liao and J. F. Hartwig, J. Am. Chem. Soc.,
C. N. Hadei, G. A. Chass, A. Lough, A. C. Hopkinson and
M. G. Organ, Chem.–Eur. J., 2006, 12, 4743–4748; (d)
K. Billingsley and S. L. Buchwald, J. Am. Chem. Soc., 2007,
129, 3358; (e) T. Kinzel, Y. Zhang and S. L. Buchwald, J.
Am. Chem. Soc., 2010, 132, 14073; (f) D. W. Robbins and
2010, 132, 11389; (b) G. Zhang, L. Zhang, M. Hu and
J. Cheng, Adv. Synth. Catal., 2011, 353, 291; (c)
P. Anbarasan, J. Neumann and M. Beller, Angew. Chem.,
Int. Ed., 2011, 50, 519; (d) J. Kim, J. Choi, K. Shin and
S. Chang, J. Am. Chem. Soc., 2012, 134, 2528; (e) Y. Luo,
Q. Wen, Z. Wu, J. Jin, P. Lu and Y. Wang, Tetrahedron,
2013, 69, 8400. For a related protocol which is reported
¨
J. F. Hartwig, Org. Lett., 2012, 14, 4266; (g) M. A. Dufert,
K. L. Billingsley and S. L. Buchwald, J. Am. Chem. Soc.,
2013, 135, 12877.
This journal is © The Royal Society of Chemistry 2016
Chem. Sci.

View Article Online
Chemical Science
Edge Article
only for boronic acid substrates, see: Z. H. Zhang and
L. S. Liebeskind, Org. Lett., 2006, 8, 4331.
22 (a) A. S. Guram, R. A. Rennels and S. L. Buchwald, Angew.
Chem., Int. Ed., 1995, 34, 1348; (b) J. Louie and
J. F. Hartwig, Tetrahedron Lett., 1995, 36, 3609; (c) A. Shar
hours. However, in order to enable quantitative LC
analysis, the generation of all reaction products at
preparative scale, isolation and full characterization
requires on the order of one to two weeks of experimental
effort per library.
and S. L. Buchwald, J. Am. Chem. Soc., 2006, 128, 8742; (d) 30 (a) J. R. Schmink, A. Bellomo and S. Berritt, Aldrichimica Acta,
N. Marion, O. Navarro, J. Mei, E. D. Stevens, N. M. Scott
and S. P. Nolan, J. Am. Chem. Soc., 2006, 128, 4101; (e)
C. T. Yang, Y. Fu, Y. B. Huang, J. Yi, Q. X. Guo and L. Liu,
Angew. Chem., Int. Ed., 2009, 48, 7398; (f) Y. Zhang,
X. Yang, Q. Yao and D. Ma, Org. Lett., 2012, 14, 3056.
23 I. P. Beletskaya and A. V. Cheprakov, Organometallics, 2012,
31, 7753.
2013, 46, 71; (b) K. D. Collins, T. Gensch and F. Glorius, Nat.
Chem., 2014, 6, 859; (c) A. Gordillo, S. Titlbach, C. Futter,
M. L. Lejkowski, E. Prasetyo, L. T. A. Rupin, T. Emmert
and S. A. Schunk, Ullmann's Encyclopedia of Industrial
Chemistry, 2014, p. 1; (d) A. B. Santanilla, E. R. Regalado,
T. Pereira, M. Shevlin, K. Bateman, L. C. Campeau,
J. Schneeweis, S. Berritt, Z.-C. Shi, P. Nantermet, Y. Liu,
R. Helmy, C. J. Welch, P. Vachal, I. W. Davies, T. Cernak
and S. D. Dreher, Science, 2015, 347, 49.
24 Q. Shen, T. Ogata and J. F. Hartwig, J. Am. Chem. Soc., 2008,
130, 6586.
25 D. Maiti, B. P. Fors, J. L. Henderson, Y. Nakamura and 31 (a) Y. Inokuma, S. Yoshioka, J. Ariyoshi, T. Arai, Y. Hitora,
S. L. Buchwald, Chem. Sci., 2011, 2, 57.
26 N. C. Bruno, M. T. Tudge and S. L. Buchwald, Chem. Sci.,
2013, 4, 916.
27 A. B. Santanilla, M. Christensen, L.-C. Campeau, I. W. Davies
and S. D. Dreher, Org. Lett., 2015, 17, 3370.
28 J. Lindley, Tetrahedron, 1984, 40, 1433.
K. Takada, S. Matsunaga, K. Rissanen and M. Fujita,
Nature, 2013, 495, 461; (b) N. Zigon, M. Hoshino,
S. Yoshioka, Y. Inokuma and M. Fujita, Angew. Chem., Int.
Ed., 2014, 54, 9033; (c) T. R. Ramadhar, S.-L. Zheng,
Y.-S. Chen and J. Clardy, Acta Crystallogr., Sect. A: Found.
Adv., 2015, 71, 46; (d) T. R. Ramadhar, S.-L. Zheng,
Y.-S. Chen and J. Clardy, Chem. Commun., 2015, 51, 11252.
29 Using the pre-plated libraries of informer compounds (18–
24 compounds per library), the experimental evaluation of 32 A. Milo, A. J. Neel, F. D. Toste and M. S. Sigman, Science,
given reaction method, including reaction set-up, 2015, 347, 737.
quantitative LC analysis and data workup, requires 3–4
a
Chem. Sci.
This journal is © The Royal Society of Chemistry 2016