Issue:
Therapeutic resistance appears to simply be a manifestation of successive selection pressures in which novel clones are selected for, or pre-existing clones within polyclonal tumors are positively selected for. For example, lineage plasticity facilitates conversion of a cancer cell that is dependent on the therapeutic target to one that becomes insensitive to its ablation.
Hypothesis:
Cannot target based on static diagnostic measure. Must develop mathematical model of clonal evolution for each patient. Need dynamic measure. 'Self-measurements' at regular intervals (ex. 12 hours) to develop trend of development - molecular profiles of individual clonal evolution and sub clonal evolution. Then treat appropriately.
Use mouse models of cancer to develop algorithms. See p53 G&D clonal evolution paper. Follow accordingly.
Self-diagnostic:
Theranos-based extraction of biomaterial (DNA, RNA, Protein) - NGS-based analysis. Extract parameters for entry into algorithm.
Algorithm:
Algorithm assumes constant principles governing change in molecular profiles. Multiple 'If -> then's'.
Ex. If 'p53 DN mutation X', then 'location/factor y susceptible to misregulation. If factor 'y' misregulated, then 'location/factor z susceptible to misregulation' etc.
Natural readers:
The immune system already has methods of 'reading' and reacting in an 'If -> then' manner. Can this be used as an inherent algorithm?
Q: Use of robust measures detectable independent of heterogeneity:
TEs are aplified as populations rather than individual elements. Amplifies 'misregulation' signals early on, even in a dilute subclonal population.
Can retrotransposition (RT) and retroelement expression (RE) signatures that correlate with particular diagnosis, prognosis, and therapeutic response.
i.e. RT and RE signature 'X' indicate small subpop. with inactivated tumor suppressor 'y'. Must be able to convey same molecular info as heterogeneity dynamics fluctuate. Might be able to with RT and RE signatures.
Must be able to translate into simple clinic-ready/ friendly assays - i.e. primer PCR-based or antibody-IHC based. etc.
To accomplish, need access to human data, mouse prospective pre-clinical models, sequencers (Illumina, PAC bio is best for repeats)
Need projects that address how to derive signatures from older input platforms to use prospectively and maximize signature power. i.e. even arrays, low coverage sequencing, etc.
Separate Q:
How are repeat expansions measured/detected? i.e. expansion of tandem satellite copy numbers - how is this determined?
Update:
Akin to machine-learning based association between epigenetic signatures and disease progression. See MeDiP-developments
Therapeutic resistance appears to simply be a manifestation of successive selection pressures in which novel clones are selected for, or pre-existing clones within polyclonal tumors are positively selected for. For example, lineage plasticity facilitates conversion of a cancer cell that is dependent on the therapeutic target to one that becomes insensitive to its ablation.
Hypothesis:
Cannot target based on static diagnostic measure. Must develop mathematical model of clonal evolution for each patient. Need dynamic measure. 'Self-measurements' at regular intervals (ex. 12 hours) to develop trend of development - molecular profiles of individual clonal evolution and sub clonal evolution. Then treat appropriately.
Use mouse models of cancer to develop algorithms. See p53 G&D clonal evolution paper. Follow accordingly.
Self-diagnostic:
Theranos-based extraction of biomaterial (DNA, RNA, Protein) - NGS-based analysis. Extract parameters for entry into algorithm.
Algorithm:
Algorithm assumes constant principles governing change in molecular profiles. Multiple 'If -> then's'.
Ex. If 'p53 DN mutation X', then 'location/factor y susceptible to misregulation. If factor 'y' misregulated, then 'location/factor z susceptible to misregulation' etc.
Natural readers:
The immune system already has methods of 'reading' and reacting in an 'If -> then' manner. Can this be used as an inherent algorithm?
Q: Use of robust measures detectable independent of heterogeneity:
TEs are aplified as populations rather than individual elements. Amplifies 'misregulation' signals early on, even in a dilute subclonal population.
Can retrotransposition (RT) and retroelement expression (RE) signatures that correlate with particular diagnosis, prognosis, and therapeutic response.
i.e. RT and RE signature 'X' indicate small subpop. with inactivated tumor suppressor 'y'. Must be able to convey same molecular info as heterogeneity dynamics fluctuate. Might be able to with RT and RE signatures.
Must be able to translate into simple clinic-ready/ friendly assays - i.e. primer PCR-based or antibody-IHC based. etc.
To accomplish, need access to human data, mouse prospective pre-clinical models, sequencers (Illumina, PAC bio is best for repeats)
Need projects that address how to derive signatures from older input platforms to use prospectively and maximize signature power. i.e. even arrays, low coverage sequencing, etc.
Separate Q:
How are repeat expansions measured/detected? i.e. expansion of tandem satellite copy numbers - how is this determined?
Update:
Akin to machine-learning based association between epigenetic signatures and disease progression. See MeDiP-developments