96
structuromic and construction of the whole genome metabolic network to Klebsiella
pneumoniae Kp13, twenty nine proteins were prioritized based on essentiality, drug
ability level, non-host homology and metabolic network analysis (Ramos
et al. 2018).
Enzyme robustness analyses assess susceptible fractions of a metabolic network
that can be used for drug targeting (Chavali et al. 2012). Constraining the metabolic
flux through partial or complete inhibition of an enzyme catalysed reaction of a
metabolic network allows the identification of the enzyme robustness (Chavali et al.
2012). The effects produced on objective flux (e.g. production of biomass, charge,
mass, and energy balanced) reveal the impact of inhibition and therefore the putative relevance of the target (Chavali et al. 2012).
For instance, the rise of multi resistant Mycobacterium tuberculosis (Koch and
Mizrahi 2018) motivated the modelling of mycobacterial metabolism. High
throughput data associated with genome scale network reconstruction with constraint based (CB) mathematical generates drug phenotype, specific growth rate,
and metabolic state predictions (Rienksma et al. 2014). Although enzyme robustness analyses may provide accurate descriptions of metabolism (Schilling and
Palsson 2000; Jamshidi and Palsson 2007; Fang et al. 2010; Raghunathan et al.
2010), the scarcity of kinetic parameters of each enzyme complicates genome
scale modelling. More recently, Chichonska and collaborators developed a computational experimental based approach to drug target interaction mapping on
kinase inhibitors (Cichonska et al. 2017). Using kernel based regression algorithm
as the prediction model, they provided a model for identification of new target
selectivity for drug repurposing applications (Cichonska et al. 2017). Even though
this approach was not developed to drugs against microbial the same strategy can
be applied.
Regarding viruses, due to the reduced number of proteins, the investigation is
focused on the host virus interaction and perturbation of metabolic subsets during
the infection. PPI (Protein-protein-interactions) data sets such as VirHostNet
(Guirimand et al. 2015), VirusMentha (Calderone et al. 2015), HCVpro (Kwofie
et al. 2011) and HHID6 (Ptak et al. 2008) allows viral infections to be exploited as
networks. Other host pathogen interaction databases and tools are shown in
Table 4.4.
A study investigating host virus interactions regulated in early stage HIV-1 replication cycle revealed 213 host cellular factors, 40 novel factors influencing the
initiation of HIV-1 DNA synthesis and 5 proteins with diverse influence at nuclear
import of viral DNA integration (Konig et al. 2008). The identification of target in
early stages in initial phases at virus infection have been used as a therapeutic targets. Small molecules can be developed in order to block early stage process in the
host virus interactions, such as attachment, penetration and uncoatting (Konig
et al. 2008).
Besides bacteria, viruses are the second main focus of host and pathogen interaction studies (Zhou et al. 2013) but metabolic network approaches have been also
applied to the identification of anti-fungal drug targets (Remmele et al. 2015;
Kaltdorf et al. 2016). Kaltdorf et al. (2016) applied the combination of different
M. Santana et al.
structuromic and construction of the whole genome metabolic network to Klebsiella
pneumoniae Kp13, twenty nine proteins were prioritized based on essentiality, drug
ability level, non-host homology and metabolic network analysis (Ramos
et al. 2018).
Enzyme robustness analyses assess susceptible fractions of a metabolic network
that can be used for drug targeting (Chavali et al. 2012). Constraining the metabolic
flux through partial or complete inhibition of an enzyme catalysed reaction of a
metabolic network allows the identification of the enzyme robustness (Chavali et al.
2012). The effects produced on objective flux (e.g. production of biomass, charge,
mass, and energy balanced) reveal the impact of inhibition and therefore the putative relevance of the target (Chavali et al. 2012).
For instance, the rise of multi resistant Mycobacterium tuberculosis (Koch and
Mizrahi 2018) motivated the modelling of mycobacterial metabolism. High
throughput data associated with genome scale network reconstruction with constraint based (CB) mathematical generates drug phenotype, specific growth rate,
and metabolic state predictions (Rienksma et al. 2014). Although enzyme robustness analyses may provide accurate descriptions of metabolism (Schilling and
Palsson 2000; Jamshidi and Palsson 2007; Fang et al. 2010; Raghunathan et al.
2010), the scarcity of kinetic parameters of each enzyme complicates genome
scale modelling. More recently, Chichonska and collaborators developed a computational experimental based approach to drug target interaction mapping on
kinase inhibitors (Cichonska et al. 2017). Using kernel based regression algorithm
as the prediction model, they provided a model for identification of new target
selectivity for drug repurposing applications (Cichonska et al. 2017). Even though
this approach was not developed to drugs against microbial the same strategy can
be applied.
Regarding viruses, due to the reduced number of proteins, the investigation is
focused on the host virus interaction and perturbation of metabolic subsets during
the infection. PPI (Protein-protein-interactions) data sets such as VirHostNet
(Guirimand et al. 2015), VirusMentha (Calderone et al. 2015), HCVpro (Kwofie
et al. 2011) and HHID6 (Ptak et al. 2008) allows viral infections to be exploited as
networks. Other host pathogen interaction databases and tools are shown in
Table 4.4.
A study investigating host virus interactions regulated in early stage HIV-1 replication cycle revealed 213 host cellular factors, 40 novel factors influencing the
initiation of HIV-1 DNA synthesis and 5 proteins with diverse influence at nuclear
import of viral DNA integration (Konig et al. 2008). The identification of target in
early stages in initial phases at virus infection have been used as a therapeutic targets. Small molecules can be developed in order to block early stage process in the
host virus interactions, such as attachment, penetration and uncoatting (Konig
et al. 2008).
Besides bacteria, viruses are the second main focus of host and pathogen interaction studies (Zhou et al. 2013) but metabolic network approaches have been also
applied to the identification of anti-fungal drug targets (Remmele et al. 2015;
Kaltdorf et al. 2016). Kaltdorf et al. (2016) applied the combination of different
M. Santana et al.