ORCID Profile
0000-0002-8454-5248
Current Organisation
San Diego State University
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: Microbiology Society
Date: 04-09-2023
Publisher: Springer Science and Business Media LLC
Date: 28-04-2017
Publisher: Springer Science and Business Media LLC
Date: 13-11-2017
Publisher: MDPI AG
Date: 21-07-2023
DOI: 10.3390/MAKE5030043
Abstract: This paper investigates the post-hoc calibration of confidence for “exploratory” machine learning classification problems. The difficulty in these problems stems from the continuing desire to push the boundaries of which categories have enough ex les to generalize from when curating datasets, and confusion regarding the validity of those categories. We argue that for such problems the “one-versus-all” approach (top-label calibration) must be used rather than the “calibrate-the-full-response-matrix” approach advocated elsewhere in the literature. We introduce and test four new algorithms designed to handle the idiosyncrasies of category-specific confidence estimation using only the test set and the final model. Chief among these methods is the use of kernel density ratios for confidence calibration including a novel algorithm for choosing the bandwidth. We test our claims and explore the limits of calibration on a bioinformatics application (PhANNs) as well as the classic MNIST benchmark. Finally, our analysis argues that post-hoc calibration should always be performed, may be performed using only the test dataset, and should be sanity-checked visually.
Publisher: MDPI AG
Date: 30-07-2023
DOI: 10.3390/V15081665
Abstract: Achromobacter species colonization of Cystic Fibrosis respiratory airways is an increasing concern. Two adult patients with Cystic Fibrosis colonized by Achromobacter xylosoxidans CF418 or Achromobacter ruhlandii CF116 experienced fatal exacerbations. Achromobacter spp. are naturally resistant to several antibiotics. Therefore, phages could be valuable as therapeutics for the control of Achromobacter. In this study, thirteen lytic phages were isolated and characterized at the morphological and genomic levels for potential future use in phage therapy. They are presented here as the Achromobacter Kumeyaay phage collection. Six distinct Achromobacter phage genome clusters were identified based on a comprehensive phylogenetic analysis of the Kumeyaay collection as well as the publicly available Achromobacter phages. The infectivity of all phages in the Kumeyaay collection was tested in 23 Achromobacter clinical isolates 78% of these isolates were lysed by at least one phage. A cryptic prophage was induced in Achromobacter xylosoxidans CF418 when infected with some of the lytic phages. This prophage genome was characterized and is presented as Achromobacter phage CF418-P1. Prophage induction during lytic phage preparation for therapy interventions require further exploration. Large-scale production of phages and removal of endotoxins using an octanol-based procedure resulted in a phage concentrate of 1 × 109 plaque-forming units per milliliter with an endotoxin concentration of 65 endotoxin units per milliliter, which is below the Food and Drugs Administration recommended maximum threshold for human administration. This study provides a comprehensive framework for the isolation, bioinformatic characterization, and safe production of phages to kill Achromobacter spp. in order to potentially manage Cystic Fibrosis (CF) pulmonary infections.
Publisher: Springer Science and Business Media LLC
Date: 14-02-2013
DOI: 10.1038/ISMEJ.2013.4
Publisher: American Society for Microbiology
Date: 10-2017
DOI: 10.1128/AAC.00954-17
Abstract: Widespread antibiotic use in clinical medicine and the livestock industry has contributed to the global spread of multidrug-resistant (MDR) bacterial pathogens, including Acinetobacter baumannii . We report on a method used to produce a personalized bacteriophage-based therapeutic treatment for a 68-year-old diabetic patient with necrotizing pancreatitis complicated by an MDR A. baumannii infection. Despite multiple antibiotic courses and efforts at percutaneous drainage of a pancreatic pseudocyst, the patient deteriorated over a 4-month period. In the absence of effective antibiotics, two laboratories identified nine different bacteriophages with lytic activity for an A. baumannii isolate from the patient. Administration of these bacteriophages intravenously and percutaneously into the abscess cavities was associated with reversal of the patient's downward clinical trajectory, clearance of the A. baumannii infection, and a return to health. The outcome of this case suggests that the methods described here for the production of bacteriophage therapeutics could be applied to similar cases and that more concerted efforts to investigate the use of therapeutic bacteriophages for MDR bacterial infections are warranted.
Publisher: MyJove Corporation
Date: 11-06-2015
DOI: 10.3791/52854
Publisher: Wiley
Date: 16-11-2015
DOI: 10.1002/PROT.24942
Publisher: American Society for Microbiology
Date: 02-2010
DOI: 10.1128/AEM.01710-09
Abstract: Eleven Bacillus isolates from the surface and subsurface waters of the Gulf of Mexico were examined for their capacity to sporulate and harbor prophages. Occurrence of sporulation in each isolate was assessed through decoyinine induction, and putative lysogens were identified by prophage induction by mitomycin C treatment. No obvious correlation between ability to sporulate and prophage induction was found. Four strains that contained inducible virus-like particles (VLPs) were shown to sporulate. Four strains did not produce spores upon induction by decoyinine but contained inducible VLPs. Two of the strains did not produce virus-like particles or sporulate significantly upon induction. Isolate B14905 had a high level of virus-like particle production and a high occurrence of sporulation and was further examined by genomic sequencing in an attempt to shed light on the relationship between sporulation and lysogeny. In silico analysis of the B14905 genome revealed four prophage-like regions, one of which was independently sequenced from a mitomycin C-induced lysate. Based on PCR and transmission electron microscopy (TEM) analysis of an induced phage lysate, one is a noninducible phage remnant, one may be a defective phage-like bacteriocin, and two were inducible prophages. One of the inducible phages contained four putative transcriptional regulators, one of which was a SinR-like regulator that may be involved in the regulation of host sporulation. Isolates that both possess the capacity to sporulate and contain temperate phage may be well adapted for survival in the oligotrophic ocean.
Publisher: Public Library of Science (PLoS)
Date: 02-11-2020
DOI: 10.1371/JOURNAL.PCBI.1007845
Abstract: For any given bacteriophage genome or phage-derived sequences in metagenomic data sets, we are unable to assign a function to 50–90% of genes, or more. Structural protein-encoding genes constitute a large fraction of the average phage genome and are among the most ergent and difficult-to-identify genes using homology-based methods. To understand the functions encoded by phages, their contributions to their environments, and to help gauge their utility as potential phage therapy agents, we have developed a new approach to classify phage ORFs into ten major classes of structural proteins or into an “other” category. The resulting tool is named PhANNs (Phage Artificial Neural Networks). We built a database of 538,213 manually curated phage protein sequences that we split into eleven subsets (10 for cross-validation, one for testing) using a novel clustering method that ensures there are no homologous proteins between sets yet maintains the maximum sequence ersity for training. An Artificial Neural Network ensemble trained on features extracted from those sets reached a test F 1 -score of 0.875 and test accuracy of 86.2%. PhANNs can rapidly classify proteins into one of the ten structural classes or, if not predicted to fall in one of the ten classes, as “other,” providing a new approach for functional annotation of phage proteins. PhANNs is open source and can be run from our web server or installed locally.
Location: United States of America
Start Date: 2019
End Date: 2024
Funder: National Institute of Diabetes and Digestive and Kidney Diseases
View Funded Activity