A data-driven approach for evaluating multi-modal therapy in traumatic brain injury

Haefeli, Jenny; Ferguson, Adam R.; Bingham, Deborah; Orr, Adrienne; Won, Seok Joon; Lam, Tina I.; Shi, Jian; Hawley, Sarah; Liu, Jialing; Swanson, Raymond A.; Massa, Stephen M.

doi:10.1038/srep42474

Cited by 18 publications

(18 citation statements)

References 57 publications

(71 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The rapid evolution of increasingly sophisticated computational modeling techniques and inexpensive data storage are beginning to converge with the efforts made by several groups to develop a global ontology and common data elements to harmonize massive amounts of preclinical data 51‐55 . By analyzing data en masse , the hope is that researchers will be able to develop hypotheses on more solid empirical ground 56 and drug developers can more accurately determine what model systems and tests to pursue; both spending less time and financial resources following independent underpowered studies (see e.g., Ref. 57).…”

Section: Discussionmentioning

confidence: 99%

Mouse model systems of autism spectrum disorder: Replicability and informatics signature

Kabitzke

Morales

et al. 2020

Genes Brain and Behavior

View full text Add to dashboard Cite

Phenotyping mouse model systems of human disease has proven to be a difficult task, with frequent poor inter‐ and intra‐laboratory replicability, particularly in behavioral domains such as social and cognitive function. However, establishing robust animal model systems with strong construct validity is of fundamental importance as they are central tools for understanding disease pathophysiology and developing therapeutics. To complete our studies of mouse model systems relevant to autism spectrum disorder (ASD), we present a replication of the main findings from our two published studies of five genetic mouse model systems of ASD. To assess the intra‐laboratory robustness of previous results, we chose the two model systems that showed the greatest phenotypic differences, the Shank3/F and Cntnap2, and repeated assessments of general health, activity and social behavior. We additionally explored all five model systems in the same framework, comparing all results obtained in this three‐yearlong effort using informatics techniques to assess commonalities and differences. Our results showed high intra‐laboratory replicability of results, even for those with effect sizes that were not particularly large, suggesting that discrepancies in the literature may be dependent on subtle but pivotal differences in testing conditions, housing enrichment, or background strains and less so on the variability of the behavioral phenotypes. The overall informatics analysis suggests that in our behavioral assays we can separate the set of tested mouse model system into two main classes that in some aspects lie on opposite ends of the behavioral spectrum, supporting the view that autism is not a unitary concept.

show abstract

Section: Discussionmentioning

confidence: 99%

Mouse model systems of autism spectrum disorder: Replicability and informatics signature

Kabitzke

Morales

et al. 2020

Genes Brain and Behavior

View full text Add to dashboard Cite

show abstract

“…The rapid evolution of increasingly sophisticated computational modeling techniques and inexpensive data storage are beginning to converge with the efforts made by several groups to develop a global ontology and common data elements to harmonize massive amounts of preclinical data (e.g., Ferguson, Nielson et al 2014, Smith, Hicks et al 2015, Lapinlampi, Melin et al 2017, Hume, Chow et al 2018, Wang, Liao et al 2018. By analyzing data en masse, the hope is that researchers will be able to develop hypotheses on more solid empirical ground (Haefeli, Ferguson et al 2017) and drug developers can more accurately determine what model systems and tests to pursue; both spending less time and financial resources following independent underpowered studies (see for example Freedman, Cockburn et al 2015).…”

Section: Discussionmentioning

confidence: 99%

Mouse Model Systems of Autism Spectrum Disorder: Replicability and Informatics Signature

Kabitzke

Morales

et al. 2019

Preprint

View full text Add to dashboard Cite

Background: Phenotyping mouse model systems of human disease has proven to be a difficult task, with frequent poor inter-and intra-laboratory replicability and translatability, particularly in behavioral domains such as social and verbal function. However, establishing robust animal model systems with strong construct validity is of fundamental importance as they are central tools for understanding disease pathophysiology and developing therapeutics. To complete our studies of mouse model systems relevant to autism spectrum disorder (ASD), we present a replication of the main findings from our two published studies comprising five genetic mouse model systems of ASD.Methods: To assess the robustness of our previous results, we chose the two model systems that showed the greatest phenotypic differences, the Shank3/F and Cntnap2, and repeated assessments of general health, activity, and social behavior. We additionally explored all five model systems in the same framework, comparing all results obtained in this three-yearlong effort using informatics techniques to look for commonalities and differences.Results: Results in the current study were very similar to our previously published results. The informatics signatures of the two model systems chosen for the replication showed that they were most distinguished by activity levels. Although the two model systems were opposite in this regard, those aspects of their social behavior not confounded by activity (vocalizations) were similar. P a g e | 2Conclusions: Our results showed high intra-laboratory replicability of results, even for those with effect sizes that were not particularly large, suggesting that discrepancies in the literature may be dependent on subtle differences in testing conditions, housing enrichment, or background strains and not so much on the variability of the behavioral phenotypes. The overall informatics analysis suggests two main classes of model systems that in some aspects lie on opposite ends of the behavioral spectrum, supporting the view that autism is not a unitary concept.4.

show abstract

“…Nevertheless, the PRS‐platform‐derived IOR can converge all the rats to have uniform tumor response. This augmented AI‐PRS platform is model independent, and can be applied toward virtually all classes of drugs, having been previously validated for indications ranging from infectious diseases to clinical immunosuppression and regenerative medicine . Future studies harness this platform to agnostically design novel drug combinations for subsequent individualization.…”

Section: Discussionmentioning

confidence: 99%

“…The broader definition of AI, particularly in the context of medicine, is the use of large data sets to train algorithms to iteratively solve for constants in these algorithms that can mediate improved image recognition for diagnostics, as well as drug discovery and development . With a few hundred experimental data points, we applied an artificial intelligence (AI)‐based neural networks approach to correlate drug–dose inputs with phenotypic outputs (tumor burden, toxicity markers) .…”

Section: Introductionmentioning

confidence: 99%

Harnessing an Artificial Intelligence Platform to Dynamically Individualize Combination Therapy for Treating Colorectal Carcinoma in a Rat Model

Chang

et al. 2019

Advanced Therapeutics

View full text Add to dashboard Cite

Designing multi-drug regimens often involves target-and synergy prediction-based drug selection, and subsequent dose escalation to achieve the maximum tolerated dose (MTD) of each drug. This approach may improve efficacy, but not the optimal efficacy and often substantially increases toxicity. Drug interactions depend on many pathways in the omics networks and further complicate the design process. The virtually infinite drug-dose parameter space cannot be reconciled using conventional approaches, which are largely based on prediction. This barrier at least partially accounts for the low response rates that are observed with conventional mono-and combinatorial chemotherapy. A combination of nonstandard therapies for colorectal cancer (AGCH: adriamycin, gemcitabine, cisplatin, and herceptin) at ¼ MTD is used to treat the rats, the tumor response rates varied in a wide range. Some of the tumor response rates are close to that of control group. This work harnesses an artificial intelligence (AI) platform that is mechanism free and can dynamically optimize combinatorial therapy in rats. The individually optimized AGCH regimen reveal starkly different drug-dose parameters, which can converge each rat toward the same and low tumor response rate. Importantly, this AI-based drug-dose optimization technology is an actionable platform, which can achieve N-of-1 therapy.

show abstract

A data-driven approach for evaluating multi-modal therapy in traumatic brain injury

Cited by 18 publications

References 57 publications

Mouse model systems of autism spectrum disorder: Replicability and informatics signature

Mouse model systems of autism spectrum disorder: Replicability and informatics signature

Mouse Model Systems of Autism Spectrum Disorder: Replicability and Informatics Signature

Harnessing an Artificial Intelligence Platform to Dynamically Individualize Combination Therapy for Treating Colorectal Carcinoma in a Rat Model

Contact Info

Product

Resources

About