1. |
- Ma, Kevin C., et al.
(author)
-
Adaptation to the cervical environment is associated with increased antibiotic susceptibility in Neisseria gonorrhoeae
- 2020
-
In: Nature Communications. - : Nature Publishing Group. - 2041-1723. ; 11:1
-
Journal article (peer-reviewed)abstract
- Neisseria gonorrhoeae is an urgent public health threat due to rapidly increasing incidence and antibiotic resistance. In contrast with the trend of increasing resistance, clinical isolates that have reverted to susceptibility regularly appear, prompting questions about which pressures compete with antibiotics to shape gonococcal evolution. Here, we used genome-wide association to identify loss-of-function (LOF) mutations in the efflux pump mtrCDE operon as a mechanism of increased antibiotic susceptibility and demonstrate that these mutations are overrepresented in cervical relative to urethral isolates. This enrichment holds true for LOF mutations in another efflux pump, farAB, and in urogenitally-adapted versus typical N. meningitidis, providing evidence for a model in which expression of these pumps in the female urogenital tract incurs a fitness cost for pathogenic Neisseria. Overall, our findings highlight the impact of integrating microbial population genomics with host metadata and demonstrate how host environmental pressures can lead to increased antibiotic susceptibility.
|
|
2. |
- Ormenisan, Alexandru-Adrian, et al.
(author)
-
Time travel and provenance for machine learning pipelines
- 2020
-
In: OpML 2020 - 2020 USENIX Conference on Operational Machine Learning. - : USENIX Association.
-
Conference paper (peer-reviewed)abstract
- Machine learning pipelines have become the defacto paradigm for productionizing machine learning applications as they clearly abstract the processing steps involved in transforming raw data into engineered features that are then used to train models. In this paper, we use a bottom-up method for capturing provenance information regarding the processing steps and artifacts produced in ML pipelines. Our approach is based on replacing traditional intrusive hooks in application code (to capture ML pipeline events) with standardized change-data-capture support in the systems involved in ML pipelines: the distributed file system, feature store, resource manager, and applications themselves. In particular, we leverage data versioning and time-travel capabilities in our feature store to show how provenance can enable model reproducibility and debugging.
|
|