Search

Abstract Explaining predictions for drug repositioning with biological knowledge graphs is a challenging problem. Graph completion methods using symbolic reasoning predict drug treatments and associated rules to generate evidence representing the therapeutic basis of the drug. Yet the vast amounts of generated paths that are biologically irrelevant or not mechanistically meaningful within the context of disease biology can limit utility. We use a reinforcement learning based knowledge graph completion model combined with an automatic filtering approach that produces the most relevant rules and biological paths explaining the predicted drug’s therapeutic connection to the disease. In this work we validate the approach against preclinical experimental data for Fragile X syndrome demonstrating strong correlation between automatically extracted paths and experimentally derived transcriptional changes of selected genes and pathways of drug predictions Sulindac and Ibudilast. Additionally, we show it reduces the number of generated paths in two case studies, 85% for Cystic fibrosis and 95% for Parkinson’s disease.

The CompactLight European consortium is designing a state-of-the-art X-ray free-electron laser driven by radiofrequency X-band technology. Rooted in experimental data on photo-injector performance in the recent literature, this study estimates analytically and numerically the performance of the CompactLight delivery system for bunch charges in the range 75–300 pC. Space-charge forces in the injector, linac transverse wakefield, and coherent synchrotron radiation in bunch compressors are all taken into account. The study confirms efficient lasing in the soft X-rays regime with pulse energies up to hundreds of microjoules at repetition rates as high as 1 kHz.

Ver PDF

This work presents ab initio calculations for the Kα spectrum of manganese (Z = 25, [Ar]3d54s2), a highly complex system due to the five open orbitals in the 3d shell. The spectrum is composed of the canonical diagram line [1s]→[2p] and shake-off satellite lines [1snl]→[2pnl] (nl∈{2s,2p,3s,3p,3d,4s}), where square brackets denote a hole state. The multiconfiguration Dirac–Hartree–Fock method with the active set approach provides the initial and final atomic wavefunctions. Results are presented as energy eigenvalue spectra for the diagram and satellite transitions. The calculated wavefunctions include over one hundred million configuration state functions and over 280,000 independent transition energies for the seven sets of spectra considered. Shake-off probabilities and Auger transition rates determine satellite intensities. The number of configuration state functions ensures highly-converged wavefunctions. Several measures of convergence demonstrate convergence in the calculated parameters. We obtain convergence of the transition energies in all eight transitions to within 0.06 eV and shake-off probabilities to within 4.5%.