Statistical detection of synergy: New methods and a comparative study.

Combination therapies are increasingly adopted as the standard of care for various diseases to improve treatment response, minimise the development of resistance and/or minimise adverse events. Therefore, synergistic combinations are screened early in the drug discovery process, in which their potential is evaluated by comparing the observed combination effect to that expected under a null model. Such methodology is implemented in the BIGL R-package which allows for a quick screening of drug combinations. We extend the meanR and maxR tests from this package by allowing non-constant variance of the responses and by extending the list of null models (Loewe, Loewe2, HSA, Bliss). These new tests are evaluated in a comprehensive simulation study under various models for additivity and synergy, various monotherapeutic dose-response models (complete, partial and incomplete responders) and various types of deviation from the constant variance assumption. In addition, the BIGL package is extended with bootstrap confidence intervals for the individual off-axis points and for the overall synergy strength, which were demonstrated to have reliable coverage and can complement the existing tests. We conclude that the differences in performance between the different null models are small and depend on the simulation scenario. As a result, the choice of null model should be driven by expert knowledge on the particular problem. Finally, we demonstrate the new features of the BIGL package and the difference between the synergy models on a real dataset from drug discovery. The BIGL package is available at CRAN (https://CRAN.R-project.org/package=BIGL) and as a Shiny app (https://synergy.openanalytics.eu/app).

Free-beam spectral self-compression at supercritical peak powers.

We demonstrate free-beam spectral self-compression of ~100-GW femtosecond laser pulses due to self-phase modulation (SPM) in a transparent dielectric. While all the earlier studies of SPM-induced spectral narrowing have been performed using optical fibers, experiments and simulations presented in this Letter show that this type of spectral transformation can be implemented as a part of a full three-dimensional field-waveform dynamics and can be extended to peak powers ∼105 times higher than the critical power of self-focusing. With a properly chosen initial chirp, spectral self-compression is accompanied by pulse compression, providing spectral-temporal mode self-compression as a whole.

Hybrid x-ray laser-plasma/laser-synchrotron facility for pump-probe studies of the extreme state of matter at NRC "Kurchatov Institute".

We developed a hybrid optical pump-x-ray probe facility based on the "Kurchatov's synchrotron radiation source" and terawatt (TW) femtosecond laser. The bright x-ray photon source is based on either synchrotron radiation [up to 6 × 1014 photons/(s mm2 mrad2 0.1% bandwidth)] or laser-plasma generators (up to 108 photons/sr/pulse). The terawatt (TW) femtosecond laser pulse initiated phase transitions and a non-stationary "extreme" state of matter, while the delayed x-ray pulse acts as a probe. The synchronization between synchrotron radiation and laser pulses is achieved at 60.3 MHz using an intelligent field-programmable gate array-based phased locked loop. The timing jitter of the system is less than 30 ps. In laser-plasma sources, the x-ray and laser pulses are automatically synchronized because they are produced by using the same laser source (TW laser system). We have reached an x-ray yield of about 106 photons/sr/pulse with 6-mJ sub-ps laser pulses and using helium as a local gas medium. Under vacuum conditions, the laser energy increase up to 40 mJ leads to the enhancement of the x-ray yield of up to 108 photons/sr/pulse. The developed hybrid facility paves the way for a new class of time-resolved x-ray optical pump-probe experiments in the time interval from femtoseconds to microseconds and the energy spectrum from 3 to 30 keV.

Malignant and benign thyroid nodule differentiation through the analysis of blood plasma with terahertz spectroscopy.

The liquid and lyophilized blood plasma of patients with benign or malignant thyroid nodules and healthy individuals were studied by terahertz (THz) time-domain spectroscopy and machine learning. The blood plasma samples from malignant nodule patients were shown to have higher absorption. The glucose concentration and miRNA-146b level were correlated with the sample's absorption at 1 THz. A two-stage ensemble algorithm was proposed for the THz spectra analysis. The first stage was based on the Support Vector Machine with a linear kernel to separate healthy and thyroid nodule participants. The second stage included additional data preprocessing by Ornstein-Uhlenbeck kernel Principal Component Analysis to separate benign and malignant thyroid nodule participants. Thus, the distinction of malignant and benign thyroid nodule patients through their lyophilized blood plasma analysis by terahertz time-domain spectroscopy and machine learning was demonstrated.

Direct nose to brain delivery of small molecules: critical analysis of data from a standardized in vivo screening model in rats.

The blood-brain barrier (BBB) is often a limiting factor for getting drugs in the brain. Bypassing the BBB by intranasal (IN), or also called nose to brain (NTB), route is an interesting and frequently investigated concept for brain drug delivery. However, despite the body of evidence for IN drug delivery in literature over the last decades, reproducibility and interpretation of animal data remain challenging. The objective of this project was to assess the feasibility and value of a standardized IN screening model in rats for the evaluation of direct brain delivery. A chemically diverse set of commercial and internal small molecules were tested in the in vivo model with different doses and/or formulations. Data were analyzed using different ways of ratio calculations: blood concentration at time of sacrifice, total exposure in blood (area under the curve, AUC) and the brain or olfactory bulb concentrations. The IN route was compared to another parenteral route to decide if there is potential direct brain transport. The results show that blood and tissue concentrations and ratios are highly variable and not always reproducible. Potential direct brain delivery was concluded for some compounds, however, sometimes depending on the analysis: using blood levels at sacrifice or AUC could lead to different conclusions. We conclude that a screening model for the evaluation of direct brain transport of small molecules is very difficult to achieve and a conclusion based on a limited number of animals with this variability is questionable.

BIGL: Biochemically Intuitive Generalized Loewe null model for prediction of the expected combined effect compatible with partial agonism and antagonism.

Clinical efficacy regularly requires the combination of drugs. For an early estimation of the clinical value of (potentially many) combinations of pharmacologic compounds during discovery, the observed combination effect is typically compared to that expected under a null model. Mechanistic accuracy of that null model is not aspired to; to the contrary, combinations that deviate favorably from the model (and thereby disprove its accuracy) are prioritized. Arguably the most popular null model is the Loewe Additivity model, which conceptually maps any assay under study to a (virtual) single-step enzymatic reaction. It is easy-to-interpret and requires no other information than the concentration-response curves of the individual compounds. However, the original Loewe model cannot accommodate concentration-response curves with different maximal responses and, by consequence, combinations of an agonist with a partial or inverse agonist. We propose an extension, named Biochemically Intuitive Generalized Loewe (BIGL), that can address different maximal responses, while preserving the biochemical underpinning and interpretability of the original Loewe model. In addition, we formulate statistical tests for detecting synergy and antagonism, which allow for detecting statistically significant greater/lesser observed combined effects than expected from the null model. Finally, we demonstrate the novel method through application to several publicly available datasets.