Engineering Papers⌕ Search

Engineering topics

Allen, Jonathan E.

Publications and source records attributed to Allen, Jonathan E..

At least 37 records · Page 2

Machine Learning Models to Predict Inhibition of the Bile Salt Export Pump

Cholestatic liver injury is frequently associated with drug inhibition of bile salt transporters, such as the bile salt export pump (BSEP). Reliable in silico models to predict BSEP inhibition directly from chemical structures would significantly reduce costs during drug discovery and could help avoid injury to patients. We report our development of classification and regression models for BSEP inhibition with substantially improved performance over previously published models. We assessed the performance effects of different methods of chemical featurization, data set partitioning, and class labeling and identified the methods producing models that generalized best to novel chemical entities.

59 BASIC BIOLOGICAL SCIENCES↗

Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning

Accurately predicting small molecule partitioning and hydrophobicity is critical in the drug discovery process. There are many heterogeneous chemical environments within a cell and entire human body. For example, drugs must be able to cross the hydrophobic cellular membrane to reach their intracellular targets, and hydrophobicity is an important driving force for drug–protein binding. Atomistic molecular dynamics (MD) simulations are routinely used to calculate free energies of small molecules binding to proteins, crossing lipid membranes, and solvation but are computationally expensive. Machine learning (ML) and empirical methods are also used throughout drug discovery but rely on experimental data, limiting the domain of applicability. We present atomistic MD simulations calculating 15,000 small molecule free energies of transfer from water to cyclohexane. This large data set is used to train ML models that predict the free energies of transfer. We show that a spatial graph neural network model achieves the highest accuracy, followed closely by a 3D-convolutional neural network, and shallow learning based on the chemical fingerprint is significantly less accurate. A mean absolute error of ~4 kJ/mol compared to the MD calculations was achieved for our best ML model. We also show that including data from the MD simulation improves the predictions, tests the transferability of each model to a diverse set of molecules, and show multitask learning improves the predictions. This work provides insight into the hydrophobicity of small molecules and ML cheminformatics modeling, and our data set will be useful for designing and testing future ML cheminformatics methods.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Binding Affinity Prediction by Pairwise Function Based on Neural Network

In this paper, we present a new approach to estimate the binding affinity from given three-dimensional poses of protein–ligand complexes. In this scheme, every protein–ligand atom pair makes an additive free-energy contribution. The sum of these pairwise contributions then gives the total binding free energy or the logarithm of the dissociation constant. The pairwise contribution is calculated by a function implemented via a neural network that takes the properties of the two atoms and their distance as input. The pairwise function is trained using a portion of the PDBbind 2018 data set. The model achieves good accuracy for affinity predictions when evaluated with PDBbind 2018 and with the CASF-2016 benchmark, comparing favorably to many scoring functions such as that of AutoDock Vina. The framework here may be extended to incorporate other factors to further improve its accuracy and power.

59 BASIC BIOLOGICAL SCIENCES↗

A novel variant of torque teno virus 7 identified in patients with Kawasaki disease

Kawasaki disease (KD), first identified in 1967, is a pediatric vasculitis of unknown etiology that has an increasing incidence in Japan and many other countries. KD can cause coronary artery aneurysms. Its epidemiological characteristics, such as seasonality and clinical picture of acute systemic inflammation with prodromal intestinal/respiratory symptoms, suggest an infectious etiology for KD. Interestingly, multiple host genotypes have been identified as predisposing factors for KD. To explore experimental methodology for identifying etiological agent(s) for KD and to optimize epidemiological study design (particularly the sample size) for future studies, we conducted a pilot study. For a 1-year period, we prospectively enrolled 11 patients with KD. To each KD patient, we assigned two control individuals (one with diarrhea and the other with respiratory infections), matched for age, sex, and season of diagnosis. During the acute phase of disease, we collected peripheral blood, nasopharyngeal aspirate, and feces. We also determined genotypes, to identify those that confer susceptibility to KD. There was no statistically significant difference in the frequency of the risk genotypes between KD patients and control subjects. We also used unbiased metagenomic sequencing to analyze these samples. Metagenomic sequencing and PCR detected torque teno virus 7 (TTV7) in two patients with KD (18%), but not in control subjects (P = 0.111). Sanger sequencing revealed that the TTV7 found in the two KD patients contained almost identical variants in nucleotide and identical changes in resulting amino acid, relative to the reference sequence. Additionally, we estimated the sample size that would be required to demonstrate a statistical correlation between TTV7 and KD. Future larger scale studies with carefully optimized metagenomic sequencing experiments and adequate sample size are warranted to further examine the association between KD and potential pathogens, including TTV7.

59 BASIC BIOLOGICAL SCIENCES↗

Predicting tumor cell line response to drug pairs with deep learning

The National Cancer Institute drug pair screening effort against 60 well-characterized human tumor cell lines (NCI-60) presents an unprecedented resource for modeling combinational drug activity. We present a computational model for predicting cell line response to a subset of drug pairs in the NCI-ALMANAC database. Based on residual neural networks for encoding features as well as predicting tumor growth, our model explains 94% of the response variance. While our best result is achieved with a combination of molecular feature types (gene expression, microRNA and proteome), we show that most of the predictive power comes from drug descriptors. To further demonstrate value in detecting anticancer therapy, we rank the drug pairs for each cell line based on model predicted combination effect and recover 80% of the top pairs with enhanced activity. We present promising results in applying deep learning to predicting combinational drug response. Our feature analysis indicates screening data involving more cell lines are needed for the models to make better use of molecular features.

59 BASIC BIOLOGICAL SCIENCES↗

Gene expression analysis of whole blood RNA from pigs infected with low and high pathogenic African swine fever viruses

African swine fever virus (ASFV) is a macrophage-tropic virus responsible for ASF, a transboundary disease that threatens swine production world-wide. Since there are no vaccines available to control ASF after an outbreak, obtaining an understanding of the virus-host interaction is important for developing new intervention strategies. In this study, a whole transcriptomic RNA-Seq method was used to characterize differentially expressed genes in pigs infected with a low pathogenic ASFV isolate, OUR T88/3 (OURT), or the highly pathogenic Georgia 2007/1 (GRG). After infection, pigs infected with OURT showed no or few clinical signs; whereas, GRG produced clinical signs consistent with acute ASF. RNA-Seq detected the expression of ASFV genes from the whole blood of the GRG, but not the OURT pigs, consistent with the pathotypes of these strains and the replication of GRG in circulating monocytes. Even though GRG and OURT possess different pathogenic properties, there was significant overlap in the most upregulated host genes. A small number of differentially expressed microRNAs were also detected in GRG and OURT pigs. These data confirm previous studies describing the response of macrophages and lymphocytes to ASFV infection, as well as reveal unique gene pathways upregulated in response to infection with GRG.

59 BASIC BIOLOGICAL SCIENCES↗

Microbial Characteristics of ISS Environmental Surfaces

The microbiome of environmental surfaces from the International Space Station were characterized in order to examine the relationship to crew and hardware maintenance. The Microbial Observatory (ISS-MO) experiment generated a microbial census of ISS environments using advanced molecular microbial community analyses along with traditional culture-based methods. Since the “omics” methodologies generated an extensive microbial census, significant insights into spaceflight-induced changes in the populations of beneficial and/or potentially harmful microbes were gained. Surface samples were collected from several ISS surface locations from three flight opportunities, and were returned to Earth via the Soyuz TMA-14M or the Space X Dragon capsule. In addition to cultivation methods, viable microbial burden, iTag-based sequencing, and metagenome analyses were carried out. The cultivable microbial bioburden differed by location and sampling event. Exploring the ISS environmental microbiome revealed presence of opportunistic pathogens and antibiotic resistant microbes. Genes involved in ATP binding cassette transporters, two component systems, and beta-lactam resistance were among a diverse set of metabolic and genetic information processing pathways. Whole genome sequencing (WGS) of 50 ISS strains exhibiting resistance to various antibiotics was carried out. The antibiotic resistant genes deduced from the WGS were compared with the resistomes generated directly from the gene pool of the environmental samples. Two unique Aspergillus fumigatus strains isolated from the ISS were characterized and compared to the experimentally established clinical isolates Af293 and CEA10. A virulence assessment in a neutrophil-deficient larval zebrafish model of invasive aspergillosis indicated that both ISSFT-021 and IF1SW-F4 were significantly more lethal compared to Af293 and CEA10. The findings from this Environmental “Omics” project should be exploited to enhance human health and well-being of a closed system. In other words, the ISS-MO research aims to "translate" findings in fundamental research into medical practice (pathogen detection) and meaningful health outcomes (countermeasure development).

Perry, Jay↗

A molecular portrait of maternal sepsis from Byzantine Troy

Pregnancy complications are poorly represented in the archeological record, despite their importance in contemporary and ancient societies. While we excavated a Byzantine cemetery in Troy, we discovered calcified abscesses among a woman’s remains. Scanning electron microscopy of the tissue revealed ‘ghost cells’, resulting from dystrophic calcification, which preserved ancient maternal, fetal and bacterial DNA of a severe infection, likely chorioamnionitis. Gardnerella vaginalis and Staphylococcus saprophyticus dominated the abscesses. Phylogenomic analyses of ancient, historical, and contemporary data showed that G. vaginalis Troy fell within contemporary genetic diversity, whereas S. saprophyticus Troy belongs to a lineage that does not appear to be commonly associated with human disease today. We speculate that the ecology of S. saprophyticus infection may have differed in the ancient world as a result of close contacts between humans and domesticated animals. Our results highlight the complex and dynamic interactions with our microbial milieu that underlie severe maternal infections.

59 BASIC BIOLOGICAL SCIENCES↗

Sendai virus intra-host population dynamics and host immunocompetence influence viral virulence during in vivo passage

In vivo serial passage of non-pathogenic viruses has been shown to lead to increased viral virulence, and although the precise mechanism(s) are not clear, it is known that both host and viral factors are associated with increased pathogenicity. Under- or overnutrition leads to a decreased or dysregulated immune response and can increase viral mutant spectrum diversity and virulence. The objective of this study was to identify the role of viral mutant spectra dynamics and host immunocompetence in the development of pathogenicity during in vivo passage. Because the nutritional status of the host has been shown to affect the development of viral virulence, the diet of animal model reflected two extremes of diets which exist in the global population, malnutrition and obesity. Sendai virus was serially passaged in groups of mice with differing nutritional status followed by transmission of the passaged virus to a second host species, guinea pigs. Viral population dynamics were characterized using deep sequence analysis and computational modeling. Histopathology, viral titer and cytokine assays were used to characterize viral virulence. Viral virulence increased with passage and the virulent phenotype persisted upon passage to a second host species. Additionally, nutritional status of mice during passage influenced the phenotype. Sequencing revealed the presence of several non-synonymous changes in the consensus sequence associated with passage, a majority of which occurred in the hemagglutinin-neuraminidase and polymerase genes, as well as the presence of persistent high frequency variants in the viral population. In particular, an N1124D change in the consensus sequences of the polymerase gene was detected by passage 10 in a majority of the animals. In vivo comparison of an 1124D plaque isolate to a clone with 1124N genotype indicated that 1124D was associated with increased virulence.

59 BASIC BIOLOGICAL SCIENCES↗

Middle East Respiratory Syndrome Coronavirus Intra-Host Populations Are Characterized by Numerous High Frequency Variants

Middle East respiratory syndrome coronavirus (MERS-CoV) is an emerging human pathogen related to SARS virus. In vitro studies indicate this virus may have a broad host range suggesting an increased pandemic potential. Genetic and epidemiological evidence indicate camels serve as a reservoir for MERS virus but the mechanism of cross species transmission is unclear and many questions remain regarding the susceptibility of humans to infection. Deep sequencing data was obtained from the nasal samples of three camels that had been experimentally infected with a human MERS-CoV isolate. A majority of the genome was covered and average coverage was greater than 12,000x depth. Although only 5 mutations were detected in the consensus sequences, 473 intrahost single nucleotide variants were identified. Lastly, many of these variants were present at high frequencies and could potentially influence viral phenotype and the sensitivity of detection assays that target these regions for primer or probe binding.

59 BASIC BIOLOGICAL SCIENCES↗

Targeted amplification for enhanced detection of biothreat agents by next-generation sequencing

Background: Historically, identification of causal agents of disease has relied heavily on the ability to culture the organism in the laboratory and/or the use of pathogen-specific antibodies or sequence-based probes. However, these methods can be limiting: Even highly sensitive PCR-based assays must be continually updated due to signature degradation as new target strains and near neighbors are sequenced. Thus, there has been a need for assays that do not suffer as greatly from these limitations and/or biases. Recent advances in library preparation technologies for Next-Generation Sequencing (NGS) are focusing on the use of targeted amplification and targeted enrichment/ capture to ensure that the most highly discriminating regions of the genomes of known targets (organism-unique regions and/or regions containing functionally important genes or phylogenetically-discriminating SNPs) will be sequenced, regardless of the complex sample background. Results: In the present study, we have assessed the feasibility of targeted sequence enhancement via amplification to facilitate detection of a bacterial pathogen present in low copy numbers in a background of human genomic material. Our results indicate that the targeted amplification of signature regions can effectively identify pathogen genomic material present in as little as 10 copies per ml in a complex sample. Importantly, the correct species and strain calls could be made in amplified samples, while this was not possible in unamplified samples. Conclusions: The results presented here demonstrate the efficacy of a targeted amplification approach to biothreat detection, using multiple highly-discriminative amplicons per biothreat organism that provide redundancy in case of variation in some primer regions. Importantly, strain level discrimination was possible at levels of 10 genome equivalents. Similar results could be obtained through use of panels focused on the identification of amplicons targeted for specific genes or SNPs instead of, or in addition to, those targeted for specific organisms (ongoing gene-targeting work to be reported later). Note that without some form of targeted enhancement, the enormous background present in complex clinical and environmental samples makes it highly unlikely that sufficient coverage of key pathogen(s) present in the sample will be achieved with current NGS technology to guarantee that the most highly discriminating regions will be sequenced.

59 BASIC BIOLOGICAL SCIENCES↗

Using populations of human and microbial genomes for organism detection in metagenomes

Identifying causative disease agents in human patients from shotgun metagenomic sequencing (SMS) presents a powerful tool to apply when other targeted diagnostics fail. Numerous technical challenges remain, however, before SMS can move beyond the role of research tool. Accurately separating the known and unknown organism content remains difficult, particularly when SMS is applied as a last resort. The true amount of human DNA that remains in a sample after screening against the human reference genome and filtering nonbiological components left from library preparation has previously been underreported. In this study, we create the most comprehensive collection of microbial and reference-free human genetic variation available in a database optimized for efficient metagenomic search by extracting sequences from GenBank and the 1000 Genomes Project. The results reveal new human sequences found in individual Human Microbiome Project (HMP) samples. Individual samples contain up to 95% human sequence, and 4% of the individual HMP samples contain 10% or more human reads. In conclusion, left unidentified, human reads can complicate and slow down further analysis and lead to inaccurately labeled microbial taxa and ultimately lead to privacy concerns as more human genome data is collected.

59 BASIC BIOLOGICAL SCIENCES↗

Metagenomic Analysis of the Airborne Environment in Urban Spaces

The organisms in aerosol microenvironments, especially densely populated urban areas, are relevant to maintenance of public health and detection of potential epidemic or biothreat agents. To examine aerosolized microorganisms in this environment, we performed sequencing on the material from an urban aerosol surveillance program. Whole metagenome sequencing was applied to DNA extracted from air filters obtained during periods from each of the four seasons. The composition of bacteria, plants, fungi, invertebrates, and viruses demonstrated distinct temporal shifts. Bacillus thuringiensis serovar kurstaki was detected in samples known to be exposed to aerosolized spores, illustrating the potential utility of this approach for identification of intentionally introduced microbial agents. Together, these data demonstrate the temporally dependent metagenomic complexity of urban aerosols and the potential of genomic analytical techniques for biosurveillance and monitoring of threats to public health.

59 BASIC BIOLOGICAL SCIENCES↗