Search

Cancer drug response prediction (DRP) models present a promising approach towards precision oncology, tailoring treatments to individual patient profiles. While deep learning (DL) methods have shown great potential in this area, models that can be successfully translated into clinical practice and shed light on the molecular mechanisms underlying treatment response will likely emerge from collaborative research efforts. This highlights the need for reusable and adaptable models that can be improved and tested by the wider scientific community. In this study, we present a scoring system for assessing the reusability of prediction DRP models, and apply it to 17 peer-reviewed DL-based DRP models. As part of the IMPROVE (Innovative Methodologies and New Data for Predictive Oncology Model Evaluation) project, which aims to develop methods for systematic evaluation and comparison DL models across scientific domains, we analyzed these 17 DRP models focusing on three key categories: software environment, code modularity, and data availability and preprocessing. While not the primary focus, we also attempted to reproduce key performance metrics to verify model behavior and adaptability. Our assessment of 17 DRP models reveals both strengths and shortcomings in model reusability. To promote rigorous practices and open-source sharing, we offer recommendations for developing and sharing prediction models. Following these recommendations can address many of the issues identified in this study, improving model reusability without adding significant burdens on researchers. This work offers the first comprehensive assessment of reusability and reproducibility across diverse DRP models, providing insights into current model sharing practices and promoting standards within the DRP and broader AI-enabled scientific research community., Comment: 12 pages, 2 figures

Molecular design based on generative models, such as variational autoencoders (VAEs), has become increasingly popular in recent years due to its efficiency for exploring high-dimensional molecular space to identify molecules with desired properties. While the efficacy of the initial model strongly depends on the training data, the sampling efficiency of the model for suggesting novel molecules with enhanced properties can be further enhanced via latent space optimization. In this paper, we propose a multi-objective latent space optimization (LSO) method that can significantly enhance the performance of generative molecular design (GMD). The proposed method adopts an iterative weighted retraining approach, where the respective weights of the molecules in the training data are determined by their Pareto efficiency. We demonstrate that our multi-objective GMD LSO method can significantly improve the performance of GMD for jointly optimizing multiple molecular properties., Comment: 23 pages, 9 figures

The effects of Co2+ doping on the structural, magnetic and dielectric properties of the multiferroic frustrated antiferromagnet Mn3TeO6 have been investigated. Ceramic samples of the solid solution series Mn3-xCoxTeO6 were prepared by a solid-state reaction route. X-ray and neutron powder diffraction and electron microscopy techniques were combined with calorimetric, dielectric and magnetic measurements to investigate the dependence of the crystal structure and physical properties on temperature and composition. It is shown that the compounds with x up to 2.4 all adopt the trigonal corundum-related structure of pure Mn3TeO6 (space group R-3) in the temperature range 5-295 K and that the lattice parameters a and c and the unit-cell volume V decrease linearly with increasing Co2+ concentration. The low- temperature magnetic susceptibility and heat capacity data evidence the antiferromagnetic ordering of all samples. The Neel temperature linearly increases with Co2+ concentration x. Curie-Weiss fits of the high temperature susceptibility indicate that the magnetic frustration decreases with x. The derived magnetic structure of Mn3TeO6 can be described as an incommensurately modulated magnetic spin state with k = [0, 0, kz] and an elliptical spin- spiral order of spins within the chains of MnO6 octahedra. With increasing Co2+ concentration the propagation vector kz changes from 0.453 (x = 0) to 0.516 (x = 2.4). The magnetic anisotropy changes as well, leading to a reorientation of the spiral-basal plane. A possible coexistence of long-range order of electrical dipoles and magnetic moments in Mn3-xCoxTeO6 is discussed., Comment: 30 pages, 13 figures

Several M3 TeO6 (M = Mn, Co, Ni, Cu) oxides order antiferromagnetically at low temperatures (near or below 60 K), while displaying interesting dielectric properties at high temperatures (ferroelectricity below 1000 K in M = Ni case). We have investigated and analyzed the structural and magnetic properties of Mn-doped Co3 TeO6 and Ni3 TeO6, which order antiferromagnetically at temperatures higher than their undoped counterparts., Comment: 3 pages, 3 figures

The complex metal oxides Ni2InSbO6 (NISO) and Ni2ScSbO6 (NSSO) have been prepared in form of polycrystalline powders by a solid state reaction route. The crystal structure and magnetic properties of the compounds were investigated using a combination of X-ray and neutron powder diffraction, electron microscopy, calorimetric and magnetic measurements. The compounds adopt a trigonal structure, space group R3, of the corundum related Ni3TeO6 (NTO) type. Only one of the octahedral Ni positions (Ni(2)) of the NTO structure was found to be occupied by In (Sc). NTO has non-centrosymmetric structure and is ferroelectric below 1000 K, dielectric and second harmonic measurements suggest that also NISO and NSSO are correspondingly ferroelectric. Magnetization measurements signified antiferromagnetic ordering below TN=60 K (NSSO) and 76 K (NISO). The magnetic structure is formed by two antiferromagnetically coupled incommensurate helices with the spiral axis along the b-axis and propagation vector k = [0, ky,0] with ky= 0.036(1) (NSSO) and ky= 0.029(1) (NISO). The observed structural and magnetic properties of NISO and NSSO are discussed and compared with those of NTO., Comment: submitted version; see journal-ref for the version revised after review, edited and published in the journal's style. 28 pages; 9 figures

A complex magnetic order of the multiferroic compound Co3TeO6 has been revealed by neutron powder diffraction studies on ceramics and crushed single crystals. The compound adopts a monoclinic structure (s.g. C2/c) in the studied temperature range 2 K - 300 K but exhibits successive antiferromagnetic transitions at low temperature. Incommensurate antiferromagnetic order with the propagation vector k1 = (0, 0.485, 0.055) sets in at 26 K. A transition to a second antiferromagnetic structure with k2 = (0, 0, 0) takes place at 21.1 K. Moreover, a transition to a commensurate antiferromagnetic structure with k3 = (0, 0.5, 0.25) occurs at 17.4 K. The magnetic structures have been determined by neutron powder diffraction using group theory analysis as a preliminary tool. Different coordinations of the Co2+ ions involved in the low-symmetry C2/c structure of Co3TeO6 render the exchange-interaction network very complex by itself. The observed magnetic phase transformations are interpreted as an evidence of competing magnetic interactions. The temperature dependent changes in the magnetic structure, derived from refinements of high-resolution neutron data, are discussed and possible mechanisms connected with the spin reorientations are described., Comment: 32 pages, 12 figures

In this work we study the interplay between the crystal structure and magnetism of the pyroarsenate \alpha-Cu_2As_2O_7 by means of magnetization, heat capacity, electron spin resonance and nuclear magnetic resonance measurements as well as density functional theory (DFT) calculations and quantum Monte Carlo (QMC) simulations. The data reveal that the magnetic Cu-O chains in the crystal structure represent a realization of a quasi-one dimensional (1D) coupled alternating spin-1/2 Heisenberg chain model with relevant pathways through non-magnetic AsO_4 tetrahedra. Owing to residual 3D interactions antiferromagnetic long range ordering at T_N\simeq10K takes place. Application of external magnetic field B along the magnetically easy axis induces the transition to a spin-flop phase at B_{SF}~1.7T (2K). The experimental data suggest that substantial quantum spin fluctuations take place at low magnetic fields in the ordered state. DFT calculations confirm the quasi-one-dimensional nature of the spin lattice, with the leading coupling J_1 within the structural dimers. QMC fits to the magnetic susceptibility evaluate J_1=164K, the weaker intrachain coupling J'_1/J_1 = 0.55, and the effective interchain coupling J_{ic1}/J_1 = 0.20., Comment: Accepted for publication in Physical Review B

The magnetic and electrical properties of Co$_3$TeO$_6$ single-crystals with corundum related structure reveal a magnetic-field induced polarization below 21 K. A sharp peak in the specific heat at \approx 18 K indicates a reconstructive-type first-order phase transition. From second-harmonic generation (SHG) measurements breaking of inversion symmetry is evident and the point-group symmetry was determined as m. The temperature and magnetic-field dependence of the magnetic and electrical polarizations are discussed in the light of the SHG results., Comment: 5 pages, 5 figures

The complex metal oxide Mn3TeO6 exhibits a corundum related structure and has been prepared both in forms of single crystals by chemical transport reactions and of polycrystalline powders by a solid state reaction route. The crystal structure and magnetic properties have been investigated using a combination of X-ray and neutron powder diffraction, electron microscopy, calorimetric and magnetic measurements. At room temperature this compound adopts a trigonal structure, space group R3 with a = 8.8679(1) {\AA}, c = 10.6727(2) {\AA}. A long-range magnetically ordered state is identified below 23 K. An unexpected feature of this magnetic structure is several types of Mn-chains. Under the action of the incommensurate magnetic propagation vector k = [0, 0, 0.4302(1)] the unique Mn site is split into two magnetically different orbits. One orbit forms a perfect helix with the spiral axis along the c-axis while the other orbit has a sine wave character along the c-axis., Comment: PDF; 16 pages, 6 figures

We investigate the influence of the cooling epoch on the formation of galaxies in a cold dark matter dominated universe. Isolated haloes, with circular speeds typical of spiral galaxies, have been selected from a low resolution numerical simulation for re-simulation at higher resolution with dark matter and gas components. The initial conditions are evolved with two smoothed particle hydrodynamics codes, TREESPH and GRAPESPH. In previous SPH simulations, strong outward transport of angular momentum has led to the formation of disc-like systems with much smaller angular momenta than observed in real disc galaxies. Here we investigate whether this problem can be circumvented if feedback processes prevent disc formation until late epochs. The results of varying the cooling epoch for each of five different haloes are analysed. When cooling and star formation occur at early times, stellar discs are destroyed during merger events and we observe similar catastrophic transport of angular momentum as seen in previous work. With cooling suppressed until z=1, discs can form by the present day with angular momenta comparable to those of observed disc galaxies. We conclude that feedback processes, which prevent gas from collapsing until late epochs, are an essential ingredient in disc galaxy formation., Comment: 19 pages, LaTeX, MNRAS format, 14 inlined and 2 external Postscript figures. Submitted to MNRAS