More Like this

1. Improving Femoral Torsion Evaluation in Infants Through Analysis of Variability and Reliability in Two-Dimensional and Three-Dimensional Imaging Measurements of the Femoral Neck Axis

   https://doi.org/10.1115/1.4067358  

   Chambers, Tamara T; Melendez, Victoria; Anderson, Andrew E; Huayamave, Victor A (August 2025, Journal of Engineering and Science in Medical Diagnostics and Therapy)

   Abstract The femoral neck axis serves as a critical parameter in evaluating hip joint health, particularly in the pediatric population. Commonly used metrics for evaluating femoral torsion, such as the femoral neck-shaft and femoral anteversion angles, rely heavily on precise definitions of the position and orientation of the femoral neck axis. Current measurement methods employing radiographs and performing two-dimensional (2D) measurements on computed tomography (CT) scans are susceptible to errors due to their reliance on reader experience and the inherent limitations in 2D measurements. We hypothesized that utilizing volumetric data would mitigate these errors and enable more accurate and reproducible measurements of the femoral neck axis using the femoral anteversion and femoral neck-shaft angles. To test this hypothesis, we analyzed a historical collection of postmortem infant femoral and pelvic bones (28 hips) aged 0 to 6.5 months, with an average estimated age of 4.68 ± 1.80 months. Our findings revealed an average neck-shaft angle of 128.00 ± 4.92 deg and femoral anteversion angle of 35.56 ± 11.68 deg across all femurs, consistent with literature values. These measurements obtained from volumetric image data were found to be repeatable and reliable compared to conventional methods. Our study suggests that the proposed methodology offers a standardized approach for obtaining repeatable and reproducible measurements, thus potentially enhancing diagnostic accuracy and clinical decision-making in assessing hip developmental conditions in pediatric patients. 

   more » « less
2. An automated protocol to construct flexibility parameters for classical forcefields: applications to metal–organic frameworks

   https://doi.org/10.1039/d4ra01859a  

   Ghanavati, Reza; Escobosa, Alma C; Manz, Thomas A (July 2024, RSC Advances)

   In this work, forcefield flexibility parameters were constructed and validated for more than 100 metal-organic frameworks (MOFs). We used atom typing to identify bond types, angle types, and dihedral types associated with bond stretches, angle bends, dihedral torsions, and other flexibility interactions. Our work used Manz’s angle-bending and dihedral-torsion model potentials. For a crystal structure containing Natoms in its unit cell, the number of independent flexibility interactions is 3(Natoms – 1). Because the number of bonds, angles, and dihedrals is normally much larger than 3(Natoms – 1), these internal coordinates are redundant. To reduce (but not eliminate) this redundancy, our protocol prunes dihedral types in a way that preserves symmetry equivalency. Next, each dihedral type is classified as non-rotatable, hindered, rotatable, or linear. We introduce a smart selection method that identifies which particular torsion modes are important for each rotatable dihedral type. Then, we computed the force constants for all flexibility interactions together via LASSO regression (i.e., regularized linear least-squares fitting) of the training dataset. LASSO automatically identifies and removes unimportant forcefield interactions. For each MOF, the reference dataset was quantum-mechanically-computed in VASP via DFT with dispersion and included: (i) finite-displacement calculations along every independent atom translation mode, (ii) geometries randomly sampled via ab initio molecular dynamics (AIMD), (iii) the optimized ground-state geometry using experimental lattice parameters, and (iv) rigid torsion scans for each rotatable dihedral type. After training, the flexibility model was validated across geometries that were not part of the training dataset. For each MOF, we computed the goodness of fit (R-squared value) and the root-mean-squared error (RMSE) separately for the training and validation datasets. We compared flexibility models with and without bond-bond cross terms. Even without cross terms, the model yielded R-squared values of 0.910 (avg across all MOFs) ± 0.018 (st. dev.) for atom-in-material forces in the validation datasets. Our SAVESTEPS protocol should find widespread applications to parameterize flexible forcefields for material datasets. We performed molecular dynamics simulations using these flexibility parameters to compute heat capacities and thermal expansion coefficients for two MOFs. 

   more » « less
3. Contribution of protein conformational heterogeneity to NMR lineshapes at cryogenic temperatures

   Xu Yi, Keith J. (October 2022, bioRxiv)

   na (Ed.)

   While low temperature NMR holds great promise for the analysis of unstable samples and for sensitizing NMR detection, spectral broadening in frozen protein samples is a common experimental challenge. One hypothesis explaining the additional linewidth is that a variety of conformations are in rapid equilibrium at room temperature and become frozen, creating an inhomogeneous distribution at cryogenic temperatures. Here we investigate conformational heterogeneity by measuring the backbone torsion angle (Ψ) in E. coli DHFR at 105K. Motivated by the particularly broad N chemical shift distribution in this and other examples, we modified an established NCCN Ψ experiment to correlate the chemical shift of Ni+1 to Ψi. With selective 15N and 13C enrichment of Ile, only the unique I60-I61 pair was expected to be detected in 13C’-15N correlation spectrum. For this unique amide we detected three different conformation basins based on dispersed chemical shifts. Backbone torsion angles Ψ were determined for each basin 114 ± 7 for the major peak, and 150 ± 8 and 164 ± 16° for the minor peak as contrasted with 118 for the X-ray crystal structure (and 118-130 for various previously reported structures). These studies support the hypothesis that inhomogeneous distributions of protein backbone torsion angles contribute to the lineshape broadening in low temperature NMR spectra. 

   more » « less
4. GeoMol: Torsional Geometric Generation of Molecular 3D Conformer Ensembles

   Ganea, Octavian-Eugen; Pattanaik, Lagnajit; Coley, Connor W.; Barzilay, Regina; Jensen, Klavs F.; Green, William H.; Jaakkola, Tommi S. (January 2021, Advances in neural information processing systems)

   Prediction of a molecule's 3D conformer ensemble from the molecular graph holds a key role in areas of cheminformatics and drug discovery. Existing generative models have several drawbacks including lack of modeling important molecular geometry elements (e.g. torsion angles), separate optimization stages prone to error accumulation, and the need for structure fine-tuning based on approximate classical force-fields or computationally expensive methods such as metadynamics with approximate quantum mechanics calculations at each geometry. We propose GeoMol--an end-to-end, non-autoregressive and SE(3)-invariant machine learning approach to generate distributions of low-energy molecular 3D conformers. Leveraging the power of message passing neural networks (MPNNs) to capture local and global graph information, we predict local atomic 3D structures and torsion angles, avoiding unnecessary over-parameterization of the geometric degrees of freedom (e.g. one angle per non-terminal bond). Such local predictions suffice both for the training loss computation, as well as for the full deterministic conformer assembly (at test time). We devise a non-adversarial optimal transport based loss function to promote diverse conformer generation. GeoMol predominantly outperforms popular open-source, commercial, or state-of-the-art machine learning (ML) models, while achieving significant speed-ups. We expect such differentiable 3D structure generators to significantly impact molecular modeling and related applications. 

   more » « less
5. No-Multiplication Deterministic Hyperdimensional Encoding for Resource-Constrained Devices

   https://doi.org/10.1109/LES.2023.3298732  

   Moghadam, Mehran Shoushtari; Aygun, Sercan; Najafi, M Hassan (December 2023, IEEE Embedded Systems Letters)

   Hyperdimensional vector processing is a nascent computing approach that mimics the brain structure and offers lightweight, robust, and efficient hardware solutions for different learning and cognitive tasks. For image recognition and classification, hyperdimensional computing (HDC) utilizes the intensity values of captured images and the positions of image pixels. Traditional HDC systems represent the intensity and positions with binary hypervectors of 1K–10K dimensions. The intensity hypervectors are cross-correlated for closer values and uncorrelated for distant values in the intensity range. The position hypervectors are pseudo-random binary vectors generated iteratively for the best classification performance. In this study, we propose a radically new approach for encoding image data in HDC systems. Position hypervectors are no longer needed by encoding pixel intensities using a deterministic approach based on quasi-random sequences. The proposed approach significantly reduces the number of operations by eliminating the position hypervectors and the multiplication operations in the HDC system. Additionally, we suggest a hybrid technique for generating hypervectors by combining two deterministic sequences, achieving higher classification accuracy. Our experimental results show up to 102× reduction in runtime and significant memory-usage savings with improved accuracy compared to a baseline HDC system with conventional hypervector encoding. 

   more » « less