Automated Spleen Injury Detection Using 3D Active Contours and Machine Learning

The spleen is one of the most frequently injured organs in blunt abdominal trauma. Computed tomography (CT) is the imaging modality of choice to assess patients with blunt spleen trauma, which may include lacerations, subcapsular or parenchymal hematomas, active hemorrhage, and vascular injuries. While computer-assisted diagnosis systems exist for other conditions assessed using CT scans, the current method to detect spleen injuries involves the manual review of scans by radiologists, which is a time-consuming and repetitive process. In this study, we propose an automated spleen injury detection method using machine learning. CT scans from patients experiencing traumatic injuries were collected from Michigan Medicine and the Crash Injury Research Engineering Network (CIREN) dataset. Ninety-nine scans of healthy and lacerated spleens were split into disjoint training and test sets, with random forest (RF), naive Bayes, SVM, k-nearest neighbors (k-NN) ensemble, and subspace discriminant ensemble models trained via 5-fold cross validation. Of these models, random forest performed the best, achieving an Area Under the receiver operating characteristic Curve (AUC) of 0.91 and an F1 score of 0.80 on the test set. These results suggest that an automated, quantitative assessment of traumatic spleen injury has the potential to enable faster triage and improve patient outcomes.     

Importance of palladium-carbon bond energies in direct arylation of polyfluorinated benzenes

Fagnou et al. reported direct arylation reactions that use palladium catalysts to couple Ar(1)-X to Ar(2)-H with the aid of a coordinated base. These reactions are particularly favourable for polyfluorinated arenes Ar(2)-H (see S. I. Gorelsky, D. Lapointe and K. Fagnou, J. Am. Chem. Soc. 2008, 130, 10848). In this paper, we show by means of a DFT analysis how the energetics and activation energies vary with fluorine substitution and examine the structures of intermediates and transition states. The reactant is modelled by Pd(OAc)(Ph)(PMe(3))(DMA) (DMA = dimethylacetamide). The sequence consists of (a) replacement of DMA by arene, (b) Concerted Deprotonation Metallation (CMD), (c) decoordination of AcOH, (d) reductive elimination of biaryl. Many of the variations are dominated by the number of fluorine substituents ortho to the C-H bond and fall into three groups labelled accordingly: Set0Fo, Set1Fo, and Set2Fo. In the first step a coordinated solvent is replaced by the arene. The arenes of Set0Fo and Set1Fo coordinate in a conventional η(2)-CH=CH mode, whereas the arenes of Set2Fo coordinate in an η(1)-CH mode assisted by an OH-C hydrogen bond from the coordinated acetate. Both the energy barriers to CMD and the product energies fall into the three typical sets with the highest barrier and highest product energy being for Set0Fo. They correlate more satisfactorily with the variations in Pd-C bond energies than with the C-H acidities. The barriers to reductive elimination from Pd(Ph)(Ar(F))(PMe(3))(AcOH) increase systematically from Set0Fo to Set2Fo as the Pd-C bond becomes stronger in a regular fashion from Set0Fo to Set2Fo. Again there is a strong correlation between the energy barriers to reductive elimination and the Pd-C bond energies. It is found overall that the key aspects of the reactions are: (a) the lowering of the energy of the CMD step by the ortho fluorine substituents, (b) the regioselective activation of C-H bonds ortho to fluorine which is also determined at the CMD step, (c) the decoordination of AcOH, which maintains the transition state for reductive elimination at low Gibbs free energy. The presence of fluorine increases the effectiveness of the reaction in the sense of points a and b via the increasing strength of the palladium-carbon bond.     

Tripyrrolidinophosphoric acid triamide as an activator in samarium diiodide reductions

The electrochemical and spectrophotometric characterization of the complex formed from samarium diiodide and 4 equiv of tripyrrolidinophosphoric acid triamide (TPPA) is presented. Kinetic studies indicate that the SmI(2)/TPPA complex possesses reactivity greater than the complex formed between samarium diiodide and 4 equiv of HMPA. Examples of the use of SmI(2)/TPPA in synthesis are presented.     

Catalytic mechanism of Golgi-resident human tyrosylprotein sulfotransferase-2: A mass spectrometry approach

Human tyrosylprotein sulfotransferases catalyze the transfer of a sulfuryl moiety from the universal sulfate donor PAPS to the hydroxyl substituent of tyrosine residues in proteins and peptides to yield tyrosine sulfated products and PAP. Tyrosine sulfation occurs in the trans-Golgi network, affecting an estimated 1% of the tyrosine residues in all secreted and membrane-bound proteins in higher order eukaryotes. In this study, an effective LC-MS-based TPST kinetics assay was developed and utilized to measure the kinetic properties of human TPST-2 and investigate its catalytic mechanism when G protein-coupled CC-chemokine receptor 8 (CCR8) peptides were used as acceptor substrates. Through initial rate kinetics, product inhibition studies, and radioactive-labeling experiments, our data strongly suggest a two-site ping-pong model for TPST-2 action. In this mechanistic model, the enzyme allows independent binding of substrates to two distinct sites, and involves the formation of a sulfated enzyme covalent intermediate. Some insights on the important amino acid residues at the catalytic site of TPST-2 and its covalent intermediate are also presented. To our knowledge, this is the first detailed study of the reaction kinetics and mechanism reported for human TPST-2 or any other Golgi-resident sulfotransferase.     

CE‐ESI‐TOF/MS for human growth hormone analysis

CE is a powerful analytical tool used to separate intact biomolecules such as proteins. The coupling of CE with TOF/MS produces a very promising method that can be used to detect and identify proteins in different matrices. This paper describes an efficient, rapid, and simple CE‐ESI‐TOF/MS procedure for the analysis of endogenous human growth hormone and recombinant human growth hormone without sample preparation. Operational factors were optimized using an experimental design, and the method was successfully applied to distinguish human growth hormone and recombinant human growth hormone in unknown samples.     

Origin of Stability in Branched Alkanes

The potential origins of stability in branched alkanes are investigated, paying close attention to two recent hypotheses: geminal steric repulsion and protobranching. All alkane isomers through C₆H₁₄ along with heptane and octane were investigated at the MPW1B95/6‐311++G(d,p) level. Their geminal steric repulsion, total steric repulsion, and orbital interactions were evaluated by using natural bond orbital analysis. All measures of steric repulsion fail to explain the stability of branched alkanes. The extra stability of branched alkanes and protobranching, in general, is tied to stabilizing geminal σ→σ* delocalization, particularly of the type that involves adjacent C? C bonds and, thus, preferentially stabilizes branched alkanes. This picture is corroborated by valence bond calculations that attribute the effect to additional ionic structures (e.g., CH₃⁺ :CH₂ :CH₃^? and CH₃:^? CH₂: CH₃⁺ for propane) that are not possible without protobranching.     

Structural study of acetogenins by tandem mass spectrometry under high and low collision energy

Collision‐induced dissociation experiments of seven annonaceous acetogenins were carried out under high and low collision energy conditions. Each compound was studied as protonated or deprotonated and lithium‐ or sodium‐ cationized molecules, using ElectroSpray Ionisation (ESI) with a hybrid linear trap/orbitrap mass spectrometer (LTQ‐Orbitrap®). The same ion species were studied with a Matrix‐Assisted Laser Desorption Ionisation (MALDI) tandem mass spectrometer in a high collision energy regime (1 or 2 keV). Although each of the techniques showed some limitations in the detection of functional groups, unambiguous structural identification of the acetogenins was obtained. MALDI ToF‐ToF has the advantage over ESI‐based methods to provide mass spectra rich in informative fragments which allows the confirmation of some functional groups position. By contrast, ESI‐LTQ‐Orbitrap® analysis has the advantage over MALDI that the mass spectra are relatively simple with only fragments close to the functional groups. However, this technique needs to carry out experiments both in negative and positive ionization modes. Copyright © 2010 John Wiley & Sons, Ltd.     

Asymmetric synthesis of (−)-fosfomycin and its trans-(1S,2S)-diastereomer using a biocatalytic reduction as the key step

Fosfomycin is a gram positive and gram negative antibiotic that contains an asymmetric epoxide. An enzyme library was screened for its ability to reduce dimethyl(1-chloro-2-oxopropyl)phosphonate to the corresponding asymmetric chlorohydrin. Homology models were built in MOE, which were shown to accurately model the enzyme–substrate complex displaying the stereoselectivity that we observed. Two enzymes, YDR368w and YHR104w, were chosen for the scale up and synthesis of fosfomycin and its trans-(1S,2S)-diastereomer.     

Sur les sous-groupes nilpotents du groupe de Cremona

Résumé. Nous décrivons les sous-groupes nilpotents non virtuellement abéliens du groupe des transformations birationnelles du plan projectif complexe: un tel groupe est d'ordre fini ou virtuellement métabélien.      

Multi-funnel optimization using Gaussian underestimation

In several applications, underestimation of functions has proven to be a helpful tool for global optimization. In protein–ligand docking problems as well as in protein structure prediction, single convex quadratic underestimators have been used to approximate the location of the global minimum point. While this approach has been successful for basin-shaped functions, it is not suitable for energy functions with more than one distinct local minimum with a large magnitude. Such functions may contain several basin-shaped components and, thus, cannot be underfitted by a single convex underestimator. In this paper, we propose using an underestimator composed of several negative Gaussian functions. Such an underestimator can be computed by solving a nonlinear programming problem, which minimizes the error between the data points and the underestimator in the L ¹ norm. Numerical results for simulated and actual docking energy functions are presented.