We introduce the concept of physics-guided transfer learning to predict the thermal conductivity of an additively manufactured short-fiber reinforced polymer (SFRP) using micro-structural characteristics extracted from tensile tests. Developing composite manufacturing digital twins for SFRP composite processes like extrusion deposition additive manufacturing (EDAM) require extensive experimental material characterization. Even the same material system printed on different EDAM systems can result in significant changes to the printed micro-structure, affecting the mechanical and transport properties. This, in turn, makes characterization efforts expensive and time-consuming. Therefore, the objective of the paper is to address this experimental bottleneck and use prior information about the material manufactured in one extrusion system to predict its properties when manufactured in another system. To enable this framework, we assume that changes in properties of the same material when manufactured in different systems arise solely due to microstructural changes. To that end, we present a Bayesian framework that can transfer thermal conductivity properties across extrusion deposition additive manufacturing systems. While we discuss the transfer of thermal conductivity properties, the development is such that the framework can be used for the transfer of other properties that depend on microstructural characteristics defined in the manufacturing process. These include the coefficient of thermal expansion, viscoelastic properties, etc. The framework begins by using thermal conductivity data of the composite printed in one extrusion system to probabilistically infer the constituent thermal properties of the fiber and the polymer. Next, by conducting limited tensile tests of the same material printed in another extrusion system, we infer its orientation tensor. Finally, the inferred constituent thermal conductivity properties and the inferred orientation tensor are coupled using a micromechanics model to predict the thermal conductivity properties of the composite printed in the second extrusion system. We experimentally verify the predictions and show that our method provides a reliable framework for transferring material properties while accounting for epistemic and aleatory uncertainties.

With the significant increase in the availability of microbial genome sequences in recent years, resistance gene-guided genome mining has emerged as a powerful approach for identifying natural products with specific bioactivities. Here, we present the use of this approach to reveal the roseopurpurins as potent inhibitors of cyclin-dependent kinases (CDKs), a class of cell cycle regulators implicated in multiple cancers. We identified a biosynthetic gene cluster (BGC) with a putative resistance gene with homology to human CDK2. Using targeted gene disruption and transcription factor overexpression in Aspergillus uvarum, and heterologous expression of the BGC in Aspergillus nidulans, we demonstrated that roseopurpurin C (1) is produced by this cluster and characterized its biosynthesis. We determined the potency, specificity, and mechanism of action of 1 as well as multiple intermediates and shunt products produced from the BGC. We show that 1 inhibits human CDK2 with a Kiapp of 44 nM, demonstrates selectivity for clinically relevant members of the CDK family, and induces G1 cell cycle arrest in HCT116 cells. Structural analysis of 1 complexed with CDK2 revealed the molecular basis of ATP-competitive inhibition.

Orphan cytotoxins are small molecules for which the mechanism of action (MoA) is either unknown or ambiguous. Unveiling the mechanism of these compounds may lead to useful tools for biological investigation and new therapeutic leads. In selected cases, the DNA mismatch repair-deficient colorectal cancer cell line, HCT116, has been used as a tool in forward genetic screens to identify compound-resistant mutations, which have ultimately led to target identification. To expand the utility of this approach, we engineered cancer cell lines with inducible mismatch repair deficits, thus providing temporal control over mutagenesis. By screening for compound resistance phenotypes in cells with low or high rates of mutagenesis, we increased both the specificity and sensitivity of identifying resistance mutations. Using this inducible mutagenesis system, we implicate targets for multiple orphan cytotoxins, including a natural product and compounds emerging from a high-throughput screen, thus providing a robust tool for future MoA studies.

Abstract This study seeks to improve upon existing methods of characterizing soft-soils for mobility simulations which use the Discrete Element Method. The study presents a novel approach for characterizing DEM properties based on Bevameter test data. The proposed method involves training a Reduced Order Model (ROM) to predict Bevameter results from DEM particle properties and using the ROM within a multi-objective optimization framework to determine the DEM properties which best fit a target Bevameter dataset. To demonstrate the efficacy of the approach, the trained ROM was used to generate DEM models meant to mimic the Bevameter response for a variety of field-tested soils. The accuracy of the resulting DEM models was evaluated using two validation methods. One in which the error in the resulting DEM properties is quantified, and one in which the error in the resulting DEM Bevameter response is quantified. The findings of this study provide compelling evidence that the proposed approach is a promising method for characterizing the properties of soil for DEM simulations using Bevameter data. The results indicate that certain enhancements are necessary for further refinement of the approach. However, once refined, this method is expected to offer a dependable and efficient means of soil model characterization.

Abstract The trade-off between computational efficiency and accuracy of various terramechanics simulation methods is a challenge that many engineers face when conducting analysis of off-road vehicle mobility. The Discrete Element Method (DEM) can capture complex phenomena within the soil, but requires substantial time and computing resources in comparison to the lower fidelity simple terramechanics methods such as the Bekker-Wong model. In this study, we propose a novel approach that seeks to strike a balance between efficiency and fidelity in simulating tire tractive capability in soft-soils. Our approach leverages DEM to determine Bekker-Wong parameters which encapsulate the bulk response of the DEM model and can predict accurate overall tractive performance. The process involves using DEM to generate soft-soil tire performance curves and using an optimization framework to determine simple terramechanics parameters, such as Bekker-Wong parameters, that best fit the DEM response. These parameters can then be used in other simulations for quantifying vehicle performance on soft-soil without the need for cumbersome DEM simulations. The results of the study demonstrate that the approach is promising with further refinement. Future work should focus on starting with more accurate DEM soil properties and scaling the method up to the full-vehicle level. If successful this work could ultimately provide a model with good predictive capability, while still allowing fast simulation times for agile design iteration and high run-count analyses.

We have characterized a full-scale transition-edge sensor (TES) microcalorimeter array as a milestone towards the demonstration of technology readiness level (TRL) 5 for the detector array for the X-ray Integral Field Unit (X-IFU) instrument on ESA's future flagship X-Ray Observatory called Athena. We fabricated a 90 mm full-scale prototype TES array and measured the properties and the performance in a newly developed platform in which up to 960 out of ∼3,200 pixels can be read out. In this paper, we report on measurements of the uniformity of the transition temperature and shape, the uniformity of the spectral energy resolutions at 7 keV and 10 keV, the thermal crosstalk for the first, the diagonal, and the second nearest neighbors, and the energy sensitivities to environment perturbations for the magnetic field, the TES bias voltage, and the heat bath temperature.

The X-ray Integral Field Unit on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Athena</i> will be subject to a cosmic-ray induced thermal background on orbit, with energy depositions into the detector wafer leading to thermal bath fluctuations. Such fluctuations have the potential to degrade energy resolution performance of the transition-edge sensor based microcalorimeter. This problem was previously studied in simulations that modeled thermal bath fluctuations induced by cosmic-ray events and evaluated the resulting energy resolution degradation due to a simulated timeline of such events. Now taking an experimental approach, we present results using a collimated Am-241 alpha particle source to deposit a known energy to specific locations on the detector wafer. Thermal pulses induced by the alpha particle energy depositions are measured at various detector pixels for several different experimental configurations, including for energy deposited into the inter-pixel structure of the wafer, as well as the frame area outside the pixel array. Further, we also test both with and without a thick backside heatsinking metallization layer that is baselined for the instrument. In each case results are compared to expectations based on the thermal model developed for the previous study.

A numerical framework was developed in this study to investigate the influence of relevant Extrusion Deposition Additive Manufacturing (EDAM) processing conditions on the final fiber orientation, inter-bead void and cross-sectional geometry of the bead laid down as the extrudate. Specifically, four key processes of the EDAM process are studied, namely: (i) flow of 90° angle turn as the extrudate exits the nozzle and is laid down on the previous layer or the substrate, (ii) flow of the bead during compaction, (iii) flow of an extrudate deposited adjacent to the previously compacted bead, and (iv) flow of the adjacent bead during compaction. The simulations utilize an anisotropic viscous flow model implemented using the smoothed particle hydrodynamics method in Abaqus and fiber orientation vectors are evolved under the assumption of affine motion. The simulation results are compared with experiments for printed bead geometries produced and this comparison was shown to validate the modeling approach as useful for predicting fiber orientation, bead geometry, and bead-to-bead contact interface geometry. The processing parameters considered are nozzle height from the substrate, the ratio of the print speed to the extrusion speed (Vb/Ve), and the bead-to-bead lateral overlap distance. A series of virtual experiments were conducted, and the following observations were noted: (i) the bead printed with a high nozzle height has significantly more fiber alignment in the printing direction than the lower nozzle height, (ii) the ratio of the print speed and extrudate speed, Vb/Ve, of greater than unity results in greater fiber alignment than for Vb/Ve <1. The developed model illuminates the roles of process parameters in determining the microstructure of the printed bead and bead geometry.