Comparing feature selection and machine learning approaches for predicting CYP2D6 methylation from genetic variation

Introduction

Pharmacogenetics currently supports clinical decision-making on the basis of a limited number of variants in a few genes and may benefit paediatric prescribing where there is a need for more precise dosing. Integrating genomic information such as methylation into pharmacogenetic models holds the potential to improve their accuracy and consequently prescribing decisions. Cytochrome P450 2D6 (CYP2D6) is a highly polymorphic gene conventionally associated with the metabolism of commonly used drugs and endogenous substrates. We thus sought to predict epigenetic loci from single nucleotide polymorphisms (SNPs) related to CYP2D6 in children from the GUSTO cohort.

Methods

Buffy coat DNA methylation was quantified using the Illumina Infinium Methylation EPIC beadchip. CpG sites associated with CYP2D6 were used as outcome variables in Linear Regression, Elastic Net and XGBoost models. We compared feature selection of SNPs from GWAS mQTLs, GTEx eQTLs and SNPs within 2 MB of the CYP2D6 gene and the impact of adding demographic data. The samples were split into training (75%) sets and test (25%) sets for validation. In Elastic Net model and XGBoost models, optimal hyperparameter search was done using 10-fold cross validation. Root Mean Square Error and R-squared values were obtained to investigate each models' performance. When GWAS was performed to determine SNPs associated with CpG sites, a total of 15 SNPs were identified where several SNPs appeared to influence multiple CpG sites.

Results

Overall, Elastic Net models of genetic features appeared to perform marginally better than heritability estimates and substantially better than Linear Regression and XGBoost models. The addition of nongenetic features appeared to improve performance for some but not all feature sets and probes. The best feature set and Machine Learning (ML) approach differed substantially between CpG sites and a number of top variables were identified for each model.

Discussion

The development of SNP-based prediction models for CYP2D6 CpG methylation in Singaporean children of varying ethnicities in this study has clinical application. With further validation, they may add to the set of tools available to improve precision medicine and pharmacogenetics-based dosing.

Transdiagnostic clustering of self-schema from self-referential judgements identifies subtypes of healthy personality and depression

Introduction

The heterogeneity of depressive and anxiety disorders complicates clinical management as it may account for differences in trajectory and treatment response. Self-schemas, which can be determined by Self-Referential Judgements (SRJs), are heterogeneous yet stable. SRJs have been used to characterize personality in the general population and shown to be prognostic in depressive and anxiety disorders.

Methods

In this study, we used SRJs from a Self-Referential Encoding Task (SRET) to identify clusters from a clinical sample of 119 patients recruited from the Institute of Mental Health presenting with depressive or anxiety symptoms and a non-clinical sample of 115 healthy adults. The generated clusters were examined in terms of most endorsed words, cross-sample correspondence, association with depressive symptoms and the Depressive Experiences Questionnaire and diagnostic category.

Results

We identify a 5-cluster solution in each sample and a 7-cluster solution in the combined sample. When perturbed, metrics such as optimum cluster number, criterion value, likelihood, DBI and CHI remained stable and cluster centers appeared stable when using BIC or ICL as criteria. Top endorsed words in clusters were meaningful across theoretical frameworks from personality, psychodynamic concepts of relatedness and self-definition, and valence in self-referential processing. The clinical clusters were labeled "Neurotic" (C1), "Extraverted" (C2), "Anxious to please" (C3), "Self-critical" (C4), "Conscientious" (C5). The non-clinical clusters were labeled "Self-confident" (N1), "Low endorsement" (N2), "Non-neurotic" (N3), "Neurotic" (N4), "High endorsement" (N5). The combined clusters were labeled "Self-confident" (NC1), "Externalising" (NC2), "Neurotic" (NC3), "Secure" (NC4), "Low endorsement" (NC5), "High endorsement" (NC6), "Self-critical" (NC7). Cluster differences were observed in endorsement of positive and negative words, latency biases, recall biases, depressive symptoms, frequency of depressive disorders and self-criticism.

Discussion

Overall, clusters endorsing more negative words tended to endorse fewer positive words, showed more negative biases in reaction time and negative recall bias, reported more severe depressive symptoms and a higher frequency of depressive disorders and more self-criticism in the clinical population. SRJ-based clustering represents a novel transdiagnostic framework for subgrouping patients with depressive and anxiety symptoms that may support the future translation of the science of self-referential processing, personality and psychodynamic concepts of self-definition to clinical applications.

Evaluation of a new antiresistic strategy to manage antibiotic resistance

OBJECTIVES

Systemic strategies for combating antimicrobial resistance (AMR) currently focus on limiting antibiotic use and have been generally insufficient in preventing the rise of AMR. Additionally, they often generate other adverse incentives, such as discouraging pharmaceutical companies from investing in research and development (R&D) of new antibiotics, further exacerbating the problem. This paper proposes a novel systemic strategy for tackling AMR, which we term "antiresistics": any intervention (whether a small molecule, genetic element, phage, or whole organism) that reduces resistance rates in pathogen populations. A prime example of an antiresistic would be a small molecule that specifically disrupts the maintenance of antibiotic resistance plasmids. Of note, an antiresistic would be expected to have a population-level effect and not necessarily be useful on a time scale relevant to individual patients.

METHODS

We developed a mathematical model to assess the impact of antiresistics on population resistance levels and calibrated it to longitudinal data was available at the country level. We also estimated potential impacts on idealized rates for introduction of new antibiotics.

RESULTS

The model shows that greater use of antiresistics allows for greater usage of existing antibiotics. This leads to an ability to maintain a constant overall rate of antibiotic efficacy with a slower rate of developing new antibiotics; alternatively, antiresistics has a positive benefit on the effective lifetime and thus profitability of antibiotics.

CONCLUSIONS

By directly reducing resistance rates, antiresistics can provide clear qualitative benefits (which may be quantitatively large) in terms of existing antibiotic efficacy, longevity,and alignment of incentives.

Are Monthly Market Returns Predictable?

We document significant persistence in the market timing performance of active individual investors, suggesting that some investors are skilled at timing. Using data on all trades by active Finnish individual investors over almost 15 years, we also show that the net purchases of skilled versus unskilled investors predict monthly market returns. Our results lend credibility to the view that market returns are predictable, without having to specify which variables active investors use to successfully time the market. (JEL G10, G11, G12, G14, G15).

Project IDentif.AI: Harnessing Artificial Intelligence to Rapidly Optimize Combination Therapy Development for Infectious Disease Intervention

In 2019/2020, the emergence of coronavirus disease 2019 (COVID‐19) resulted in rapid increases in infection rates as well as patient mortality. Treatment options addressing COVID‐19 included drug repurposing, investigational therapies such as remdesivir, and vaccine development. Combination therapy based on drug repurposing is among the most widely pursued of these efforts. Multi‐drug regimens are traditionally designed by selecting drugs based on their mechanism of action. This is followed by dose‐finding to achieve drug synergy. This approach is widely‐used for drug development and repurposing. Realizing synergistic combinations, however, is a substantially different outcome compared to globally optimizing combination therapy, which realizes the best possible treatment outcome by a set of candidate therapies and doses toward a disease indication. To address this challenge, the results of Project IDentif.AI (Identifying Infectious Disease Combination Therapy with Artificial Intelligence) are reported. An AI‐based platform is used to interrogate a massive 12 drug/dose parameter space, rapidly identifying actionable combination therapies that optimally inhibit A549 lung cell infection by vesicular stomatitis virus within three days of project start. Importantly, a sevenfold difference in efficacy is observed between the top‐ranked combination being optimally and sub‐optimally dosed, demonstrating the critical importance of ideal drug and dose identification. This platform is disease indication and disease mechanism‐agnostic, and potentially applicable to the systematic N‐of‐1 and population‐wide design of highly efficacious and tolerable clinical regimens. This work also discusses key factors ranging from healthcare economics to global health policy that may serve to drive the broader deployment of this platform to address COVID‐19 and future pandemics.