Comparative analysis of machine learning approaches to classify tumor mutation burden in lung adenocarcinoma using histopathology images.

Both histologic subtypes and tumor mutation burden (TMB) represent important biomarkers in lung cancer, with implications for patient prognosis and treatment decisions. Typically, TMB is evaluated by comprehensive genomic profiling but this requires use of finite tissue specimens and costly, time-consuming laboratory processes. Histologic subtype classification represents an established component of lung adenocarcinoma histopathology, but can be challenging and is associated with substantial inter-pathologist variability. Here we developed a deep learning system to both classify histologic patterns in lung adenocarcinoma and predict TMB status using de-identified Hematoxylin and Eosin (H&E) stained whole slide images. We first trained a convolutional neural network to map histologic features across whole slide images of lung cancer resection specimens. On evaluation using an external data source, this model achieved patch-level area under the receiver operating characteristic curve (AUC) of 0.78-0.98 across nine histologic features. We then integrated the output of this model with clinico-demographic data to develop an interpretable model for TMB classification. The resulting end-to-end system was evaluated on 172 held out cases from TCGA, achieving an AUC of 0.71 (95% CI 0.63-0.80). The benefit of using histologic features in predicting TMB is highlighted by the significant improvement this approach offers over using the clinical features alone (AUC of 0.63 [95% CI 0.53-0.72], p = 0.002). Furthermore, we found that our histologic subtype-based approach achieved performance similar to that of a weakly supervised approach (AUC of 0.72 [95% CI 0.64-0.80]). Together these results underscore that incorporating histologic patterns in biomarker prediction for lung cancer provides informative signals, and that interpretable approaches utilizing these patterns perform comparably with less interpretable, weakly supervised approaches.

Intramolecular energies of the cytotoxic protein CagA of Helicobacter pylori as a possible descriptor of strains' pathogenicity level.

The Helicobacter pylori cytotoxin-associated gene A (CagA) is known for causing gastroduodenal diseases, such as atrophic gastritis and peptic ulcerations. Furthermore Helicobacter pylori CagA positive strains has been reported as one of the main risk factors for gastric cancer (Parsonnet et al., 1997). Structural variations in the CagA structure can alter its affinity with the host proteins, inducing differences in the pathogenicity of H. pylori. CagA N-terminal region is characterized for be conserved among all H. pylori strains since the C-terminal region is characterized by an intrinsically disorder behavior. We generated complete structural models of CagA using different conformations of the C-terminal region for two H. pylori strains. These models contain the same EPIYA (ABC1C2) motifs but different level of pathogenicity: gastric cancer and duodenal ulcer. Using these structural models we evaluated the pathogenicity level of the H. pylori strain, based on the affinity of the interaction with SHP-2 and Grb2 receptors and on the number of interactions with the EPIYA motif. We found that the main differences in the interaction was due to the contributions of certain types of energies from each strain and not from the total energy of the molecule. Specifically, the electrostatic energy, helix dipole energy, Wander Waals clashes, torsional clash, backbone clash and cis bond energy allowed a separation between severe and mild pathology for the interaction of only CagA with SHP2.

Multicenter Evaluation of the Portrait Staph ID/R Blood Culture Panel for Rapid Identification of Staphylococci and Detection of the mecA Gene.

Bloodstream infections are a leading cause of morbidity and mortality in the United States and are associated with increased health care costs. We evaluated the Portrait Staph ID/R blood culture panel (BCP) multiplex PCR assay (Great Basin Scientific, Salt Lake City, UT) for the rapid and simultaneous identification (ID) of Staphylococcus aureus, Staphylococcus lugdunensis, and Staphylococcus species to the genus level and the detection of the mecA gene directly from a positive blood culture bottle. A total of 765 Bactec bottles demonstrating Gram-positive cocci in singles or clusters were tested during the prospective trial at 3 clinical sites. The Portrait Staph ID/R BCP results were compared with results from conventional biochemical and cefoxitin disk methods performed at an independent laboratory. Discordant ID and mecA results were resolved by rpoB gene sequencing and mecA gene sequencing, respectively. A total of 658 Staphylococcus species isolates (S. aureus, 211 isolates; S. lugdunensis, 3 isolates; and Staphylococcus spp., 444 isolates) were recovered from monomicrobial and 33 polymicrobial blood cultures. After discrepant analysis, the overall ratios of Portrait Staph ID/R BCP positive percent agreement and negative percent agreement were 99.4%/99.9% for Staphylococcus ID and 99.7%/99.2% for mecA detection.

Peptides design based on transmembrane Escherichia coli's OmpA protein through molecular dynamics simulations in water-dodecane interfaces.

Recent research efforts have focused on the production of environmentally nonthreatening products, including identifying biosurfactants that can replace conventional surfactants. In order to utilize biosurfactants in different industries such as cosmetic, food or petroleum, it is necessary to understand the underpinnings behind the interactions that could take place for biosurfactants which display potential for interface activity. This work aimed to use molecular dynamics simulations to understand the interactions of rationally obtained peptide sequences from the original sequence of the OmpA gene in Escherichia coli, based on the free energy change (ΔG) during peptide insertion at the water-dodecane interface. Seventeen OmpA-based peptide sequences were selected and analyzed based on their hydropathy index profiles. We found that free energy change due to Columbic interactions and SASA (ΔGCoul/SASA), total free energy change and MW (ΔG/MW), and free energy change due to Coulombic and van der Waals interactions (ΔGCoul/ΔGvdW) ratios could provide a better understating in the contribution of the free energy decrease at the interface. The results indicated that the peptide sequences GKNHDTGVSPVFA and THENQLGAGAFG display biosurfactant potential based on low ΔG per square nanometer, high ΔGCoul/ΔGvdW ratio, clearly defined moieties along its hydrophobic surface and sequence, and the presence of charged residues in the polar head. Clearly defined moieties and SASA were determinant for electrostatic interactions between oil-water interfaces. Experimental validations exhibited that the emulsions prepared remained stable between 3 and 27h, respectively. Even though the peptide GKNHDTGVSPVFA displays strong interactions at the interface, stabilization times showed that the peptide THENQLGAGAFG exhibited the best performance suggesting that the stability can be better described by kinetic rather than thermodynamic criteria once the emulsion is formed.

Comparison of MDCK-MDR1 and Caco-2 cell based permeability assays for anti-malarial drug screening and drug investigations.

INTRODUCTION: Malaria is a major health concern and affects over 300million people a year. Accordingly, there is an urgent need for new efficacious anti-malarial drugs. A major challenge in developing new anti-malarial drugs is to design active molecules that have preferable drug-like characteristics. These "drug-like" characteristics include physiochemical properties that affect drug absorption, distribution, metabolism, and excretion (ADME). Compounds with poor ADME profiles will likely fail in vivo due to poor pharmacokinetics and/or other drug delivery related issues. There have been numerous assays developed in order to pre-screen compounds that would likely fail in further development due to poor absorption properties including PAMPA, Caco-2, and MDCK permeability assays. METHODS: The use of cell-based permeability assays such as Caco-2 and MDCK serve as surrogate indicators of drug absorption and transport, with the two approaches often used interchangeably. We sought to evaluate both approaches in support of anti-malarial drug development. Accordingly, a comparison of both assays was conducted utilizing apparent permeability coefficient (Papp) values determined from liquid chromatography/tandem mass spectrometry (LC-MS) analyses. RESULTS: Both Caco-2 and MDCK permeability assays produced similar Papp results for potential anti-malarial compounds with low and medium permeability. Differences were observed for compounds with high permeability and compounds that were P-gp substrates. Additionally, the utility of MDCK-MDR1 permeability measurements was demonstrated in probing the role of P-glycoprotein transport in Primaquine-Chloroquine drug-drug interactions in comparison with in vivo pharmacokinetic changes. DISCUSSION: This study provides an in-depth comparison of the Caco-2 and MDCK-MDR1 cell based permeability assays and illustrates the utility of cell-based permeability assays in anti-malarial drug screening/development in regard to understanding transporter mediated changes in drug absorption/distribution.

Molecular dynamics approach to investigate the coupling of the hydrophilic-lipophilic balance with the configuration distribution function in biosurfactant-based emulsions.

Emulsion stability has been characterized by macroscopic variables such as the hydrophilic-lipophilic balance, with the aim being to predict the surfactant properties of molecules. Nevertheless, this parameter does not take the topology of the molecule into account, as it only considers its lipophilic degree. On the other hand, the classical Derjaguin-Landau-Verwey-Overbeek approach (based on the continuum model), which has been widely utilized to evaluate the stabilities of colloids, polymers, and surfactants, takes some bulk macroscopic parameters such as the shear viscosity coefficient and the dielectric permittivity into account. In the work reported here, molecular dynamics simulations were used to elucidate the mechanism of layer formation and micellar structure for different combinations of valine-aspartic acid peptides in dodecane-water emulsions, as well as their associations with the hydrophilic-lipophilic balance. The peptide-dodecane radial distribution function showed that the first peak intensity was inversely correlated with the hydrophilic-lipophilic balance; moreover, the oscillatory structural forces became increasingly prominent when the hydrophilic-lipophilic balance was decreased. Our results seem to indicate that the radial distribution function could be utilized to evaluate the stabilities of emulsions of peptides via molecular simulations.