Publications

In this study of victims of sexual abuse, the aim was to investigate the role of perceived social support and abuse characteristics in self-reported insomnia, nightmare frequency, and nightmare distress. Four hundred sixty Norwegian victims of sexual abuse completed a questionnaire assessing perceived social support, abuse characteristics, insomnia, nightmare frequency, and nightmare distress. Results show that higher levels of perceived social support were related to lower scores on all symptom outcome measures. Abuse involving oral, genital, or anal penetration was related to more insomnia symptoms. Longer duration of abuse and threatening conducted by the perpetrator were related to higher nightmare frequency, while threats and abuse involving penetration were related to higher degrees of distress associated with nightmares. In conclusion, the present study provides preliminary data indicating that perceived social support may affect the nature of sleep difficulties in sexual abuse victims. Also, more severe forms of sexual abuse are related to higher levels of sleep difficulties.

Abstract Building chemical structures of complexity and functionality approaching the level of biological systems is an ongoing challenge. A general synthetic strategy is proposed by which progressive levels of complexity are achieved through the building block approach whereby molecularly defined constructs at one level serve as constituent units of the next level, all being linked through strong bonds—”augmented reticular chemistry”. Specifically, current knowledge of linking metal complexes and organic molecules into reticular frameworks is applied here to linking the crystals of these frameworks into supercrystals (superframeworks). This strategy allows for the molecular control exercised on the molecular regime to be translated into higher augmentation levels to produce systems capable of dynamics and complex functionality far exceeding current materials.

Abstract Graph deep learning models, which incorporate a natural inductive bias for atomic structures, are of immense interest in materials science and chemistry. Here, we introduce the Materials Graph Library (MatGL), an open-source graph deep learning library for materials science and chemistry. Built on top of the popular Deep Graph Library (DGL) and Python Materials Genomics (Pymatgen) packages, MatGL is designed to be an extensible “batteries-included” library for developing advanced model architectures for materials property predictions and interatomic potentials. At present, MatGL has efficient implementations for both invariant and equivariant graph deep learning models, including the Materials 3-body Graph Network (M3GNet), MatErials Graph Network (MEGNet), Crystal Hamiltonian Graph Network (CHGNet), TensorNet and SO3Net architectures. MatGL also provides several pre-trained foundation potentials (FPs) with coverage of the entire periodic table, and property prediction models for out-of-box usage, benchmarking and fine-tuning. Finally, MatGL integrates with PyTorch Lightning to enable efficient model training.

Terpenes constitute the largest class of natural products with over 55,000 compounds with versatile applications including drugs and biofuels. Introducing structural modifications to terpenes through metabolic engineering is an efficient and sustainable way to improve their properties. Here, we report the optimization of the lepidopteran mevalonate (LMVA) pathway towards the efficient production of isopentenyl pyrophosphate (IPP) analogs as terpene precursors. First, we linked the LMVA pathway to NudB, a promiscuous phosphatase, resulting in the production of the six-carbon analog of 3-methyl-3-buten-1-ol (isoprenol), 3-ethyl-3-buten-1-ol (C6-isoprenol). Using C6-isoprenol as the final product, we then engineered the LMVA pathway by redirecting its upstream portion from a thiolase-dependent pathway to a beta-oxidation pathway. The beta-oxidation LMVA pathway transforms valeric acid, a platform chemical that can be produced from biomass, into C6-isoprenol at a titer of 110.3 mg/L, improved from 5.5 mg/L by the thiolase LMVA pathway, which used propionic acid as a feedstock. Knockout of the E. coli endogenous thiolase genes further improved the C6-isoprenol titer to 390 mg/L, implying efficient production of homo isopentenyl pyrophosphate (HIPP). The beta-oxidation LMVA-NudB pathway also converts butanoic acid and hexanoic acid into isoprenol and isoprenol's seven-carbon analog, 3-propyl-3-buten-1-ol (C7-isoprenol), respectively, suggesting the beta-oxidation LMVA pathway produces IPP and C7-IPP from the corresponding fatty acids. Fuel property tests revealed the longer chain isoprenol analogs have lower water solubilities, similar or higher energy densities, and comparable research octane number (RON) boosting effects to isopentenols. This work not only optimizes the LMVA pathway, setting the basis for homoterpene biosynthesis to expand terpene chemical space, but provides an efficient pathway to produce isoprenol analogs as next-generation biofuels from sustainable feedstocks.

An entry from the Cambridge Structural Database, the world’s repository for small molecule crystal structures. The entry contains experimental data from a crystal diffraction study. The deposited dataset for this entry is freely available from the CCDC and typically includes 3D coordinates, cell parameters, space group, experimental conditions and quality measures.

We present photometry and spectroscopy of SN2013fs and SN2013fr in the first\n100 days post-explosion. Both objects showed transient, relatively narrow\nH$\\alpha$ emission lines characteristic of SNeIIn, but later resembled normal\nSNeII-P or SNeII-L, indicative of fleeting interaction with circumstellar\nmaterial (CSM). SN2013fs was discovered within 8hr of explosion. Its light\ncurve exhibits a plateau, with spectra revealing strong CSM interaction at\nearly times. It is a less luminous version of the transitional SNIIn PTF11iqb,\nfurther demonstrating a continuum of CSM interaction intensity between SNeII-P\nand IIn. It requires dense CSM within 6.5$\\times$10$^{14}$~cm of the\nprogenitor, from a phase of advanced pre-SN mass loss shortly before explosion.\nSpectropolarimetry of SN2013fs shows little continuum polarization, but\nnoticeable line polarization during the plateau phase. SN2013fr morphed from a\nSNIIn at early times to a SNII-L. After the first epoch its narrow lines\nprobably arose from host-galaxy emission, but the bright, narrow H$\\alpha$\nemission at early times may be intrinsic. As for SN2013fs, this would point to\na short-lived phase of strong CSM interaction if proven to be intrinsic,\nsuggesting a continuum between SNeIIn and II-L. It is a low-velocity SNII-L,\nlike SN2009kr but more luminous. SN2013fr also developed an IR excess at later\ntimes, due to warm CSM dust that require a more sustained phase of strong\npre-SN mass loss.\n

Bioinspired architectures are effective in enhancing the mechanical properties of materials, yet are difficult to construct in metallic systems. The structure-property relationships of bioinspired metallic composites also remain unclear. Here, Mg-Ti composites were fabricated by pressureless infiltrating pure Mg melt into three-dimensional (3-D) printed Ti-6Al-4V scaffolds. The result was composite materials where the constituents are continuous, mutually interpenetrated in 3-D space and exhibit specific spatial arrangements with bioinspired brick-and-mortar, Bouligand, and crossed-lamellar architectures. These architectures promote effective stress transfer, delocalize damage and arrest cracking, thereby bestowing improved strength and ductility than composites with discrete reinforcements. Additionally, they activate a series of extrinsic toughening mechanisms, including crack deflection/twist and uncracked-ligament bridging, which enable crack-tip shielding from the applied stress and lead to “Γ”-shaped rising fracture resistance R-curves. Quantitative relationships were established for the stiffness and strengths of the composites by adapting classical laminate theory to incorporate their architectural characteristics.

LLM-based judges have emerged as a scalable alternative to human evaluation and are increasingly used to assess, compare, and improve models. However, the reliability of LLM-based judges themselves is rarely scrutinized. As LLMs become more advanced, their responses grow more sophisticated, requiring stronger judges to evaluate them. Existing benchmarks primarily focus on a judge's alignment with human preferences, but often fail to account for more challenging tasks where crowdsourced human preference is a poor indicator of factual and logical correctness. To address this, we propose a novel evaluation framework to objectively evaluate LLM-based judges. Based on this framework, we propose JudgeBench, a benchmark for evaluating LLM-based judges on challenging response pairs spanning knowledge, reasoning, math, and coding. JudgeBench leverages a novel pipeline for converting existing difficult datasets into challenging response pairs with preference labels reflecting objective correctness. Our comprehensive evaluation on a collection of prompted judges, fine-tuned judges, multi-agent judges, and reward models shows that JudgeBench poses a significantly greater challenge than previous benchmarks, with many strong models (e.g., GPT-4o) performing just slightly better than random guessing. Overall, JudgeBench offers a reliable platform for assessing increasingly advanced LLM-based judges. Data and code are available at https://github.com/ScalerLab/JudgeBench.

Abstract The identification and characterization of defects, on the molecular level, in metal‐organic frameworks (MOFs) remain a challenge. With the extensive use of single‐crystal X‐ray diffraction (SXRD), the missing linker defects in the zirconium‐based MOF UiO‐66, Zr 6 O 4 (OH) 4 (C 8 H 4 O 4 ) 6 , have been identified as water molecules coordinated directly to the zirconium centers. Charge balancing is achieved by hydroxide anions, which are hydrogen bonded within the pores of the framework. Furthermore, the precise nature of the defects and their concentration can be manipulated by altering the starting materials, synthesis conditions, and post‐synthetic modifications.