166
5 Future Possibilities
Further expansion of research looking at the antibacterial actions of molecular
complexes including charge transfer complexes with two (or more) components
is likely to be another fertile research area, as long as such complexes stay intact
in vivo. Work is going on currently in new drug discovery on non-covalently linked
dimers, as well as covalently linked compounds, for protein target interactions. Interlocking ring compounds (catenanes) and rotaxanes and pseudo or quasi-systems are
also being explored in drug design particularly antibacterial design. A good recent
example is the work of Wu and colleagues who have described an intriguing strongly
bactericidal heterometallic triangular necklace with Cu(I) and Pt(II) centres and nine
positive charges. This complex was active against a range of bacterial pathogens
including drug resistant strains of Pseudomonas aeruginosa, Escherichia coli and
Staphylococcus aureus. Multi-targeting was evident with this complex resulting in
DNA cleavage as well as disruption of the cell wall/plasma membrane (Wu et al.
2020).
5.3.1 In Silico Advances
The use of artificial intelligence (AI) technology by medicinal chemists to expedite
the drug discovery process is advancing (Chen and Engkvist 2019) and further refinements are likely to lead to major developments in this area in the future including
with respect to multiply active antibacterials. AI has the potential to come up with
radically new designs for such agents.
Mapping biologically active chemical space is a continuing challenge but may be
met in part by the use of AI as in the new platform called ASPIRE for pre-clinical drug
discovery with the integration of automated synthetic chemistry with high throughput
biology and artificial intelligence capabilities (Sittampalam et al. 2019). Planning
syntheses will also be expedited using AI (Segler et al. 2018). The need to diversify by
expanding the range of synthetic reactions used including developing new ones has
been discussed by Brown and Boström (2016) and updated in 2018 by Boström
et al. (2018). These broadened synthetic capabilities are likely to be required to
implement novel multi-targeting antibacterial structural designs, as well as in the
synthesis of more diverse chemical libraries for screening and then in post-screening
optimization processes. As alluded to in different parts of this book the development
of new reactions will also be assisted by thinking well outside the square in terms
of what might be possible and involve advanced AI techniques in advancing the
thoughts to specific reaction combinations.
The key to the future will revolve around synergistic interactions between human
thinking and artificial intelligence across the full gamut of the antibacterial discovery
and development paradigm including synthetic schemes, as well as toxicity and
other ADME predictions (e.g., via the Centaur platform—W. van Hoorn, Exscientia
and Centaur Chemist™; and comment by Mullard 2017; Luechtefeld et al. 2018).
Awareness, though, of human biases in selecting appropriate algorithms in the AI
process will be important and ways to reduce it might be drawn from outside the
5 Future Possibilities
Further expansion of research looking at the antibacterial actions of molecular
complexes including charge transfer complexes with two (or more) components
is likely to be another fertile research area, as long as such complexes stay intact
in vivo. Work is going on currently in new drug discovery on non-covalently linked
dimers, as well as covalently linked compounds, for protein target interactions. Interlocking ring compounds (catenanes) and rotaxanes and pseudo or quasi-systems are
also being explored in drug design particularly antibacterial design. A good recent
example is the work of Wu and colleagues who have described an intriguing strongly
bactericidal heterometallic triangular necklace with Cu(I) and Pt(II) centres and nine
positive charges. This complex was active against a range of bacterial pathogens
including drug resistant strains of Pseudomonas aeruginosa, Escherichia coli and
Staphylococcus aureus. Multi-targeting was evident with this complex resulting in
DNA cleavage as well as disruption of the cell wall/plasma membrane (Wu et al.
2020).
5.3.1 In Silico Advances
The use of artificial intelligence (AI) technology by medicinal chemists to expedite
the drug discovery process is advancing (Chen and Engkvist 2019) and further refinements are likely to lead to major developments in this area in the future including
with respect to multiply active antibacterials. AI has the potential to come up with
radically new designs for such agents.
Mapping biologically active chemical space is a continuing challenge but may be
met in part by the use of AI as in the new platform called ASPIRE for pre-clinical drug
discovery with the integration of automated synthetic chemistry with high throughput
biology and artificial intelligence capabilities (Sittampalam et al. 2019). Planning
syntheses will also be expedited using AI (Segler et al. 2018). The need to diversify by
expanding the range of synthetic reactions used including developing new ones has
been discussed by Brown and Boström (2016) and updated in 2018 by Boström
et al. (2018). These broadened synthetic capabilities are likely to be required to
implement novel multi-targeting antibacterial structural designs, as well as in the
synthesis of more diverse chemical libraries for screening and then in post-screening
optimization processes. As alluded to in different parts of this book the development
of new reactions will also be assisted by thinking well outside the square in terms
of what might be possible and involve advanced AI techniques in advancing the
thoughts to specific reaction combinations.
The key to the future will revolve around synergistic interactions between human
thinking and artificial intelligence across the full gamut of the antibacterial discovery
and development paradigm including synthetic schemes, as well as toxicity and
other ADME predictions (e.g., via the Centaur platform—W. van Hoorn, Exscientia
and Centaur Chemist™; and comment by Mullard 2017; Luechtefeld et al. 2018).
Awareness, though, of human biases in selecting appropriate algorithms in the AI
process will be important and ways to reduce it might be drawn from outside the