Matching Items (2)
Filtering by
- Creators: Marshall, Pamela
- Creators: School of Mathematical and Statistical Sciences
- Resource Type: Text
Description
I, Christopher Negrich, am the sole author of this paper, but the tools described were designed in collaboration with Andrew Hoetker. ConstrictR (constrictor) and ConstrictPy are an R package and python tool designed together. ConstrictPy implements the functions and methods defined in ConstrictR and applies data handling, data parsing, input/output (I/O), and a user interface to increase usability. ConstrictR implements a variety of common data analysis methods used for statistical and subnetwork analysis. The majority of these methods are inspired by Lionel Guidi's 2016 paper, Plankton networks driving carbon export in the oligotrophic ocean. Additional methods were added to expand functionality, usability, and applicability to different areas of data science. Both ConstrictR and ConstrictPy are currently publicly available and usable, however, they are both ongoing projects. ConstrictR is available at github.com/cnegrich and ConstrictPy is available at github.com/ahoetker. Currently, ConstrictR has implemented functions for descriptive statistics, correlation, covariance, rank, sparsity, and weighted correlation network analysis with clustering, centrality, profiling, error handling, and data parsing methods to be released soon. ConstrictPy has fully implemented and integrated the features in ConstrictR as well as created functions for I/O and conversion between pandas and R data frames with a full feature user interface to be released soon. Both ConstrictR and ConstrictPy are designed to work with minimal dependencies and maximum available information on the algorithms implemented. As a result, ConstrictR is only dependent on base R (v3.4.4) functions with no libraries imported. ConstrictPy is dependent upon only pandas, Rpy2, and ConstrictR. This was done to increase longevity and independence of these tools. Additionally, all mathematical information is documented alongside the code, increasing the available information on how these tools function. Although neither tool is in its final version, this paper documents the code, mathematics, and instructions for use, in addition to plans for future work, for of the current versions of ConstrictR (v0.0.1) and ConstrictPy (v0.0.1).
ContributorsNegrich, Christopher Alec (Author) / Can, Huansheng (Thesis director) / Hansford, Dianne (Committee member) / School of Mathematical and Statistical Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2018-05
Description
The use of DNA testing has been focused primarily on biological samples such as blood or saliva found at crime scenes. These types of evidence in the forensic field are sometimes difficult to come by, especially when there is no body to find to verify things such as identity or status of a person. In the case of the burial of a body, they can be remote and relocated multiple times depending on each situation. Clandestine burials are not uncommon especially in the Arizona desert by the United States and Mexico border. Since there is no physical body to find the next best avenue to finding a clandestine burial is through search teams which can take weeks to months or other expensive technology such as ground penetrating radar (GPR). A new more interesting avenue to search for bodies is using the most found material–soil. Technology has allowed the possibility of using soil DNA microbiome testing initially to study the varieties of microbes that compose in soil. Microbiomes are unique and plentiful and essentially inescapable as humans are hosts of millions of them. The idea of a microbiome footprint at a crime scene seems out of reach considering the millions of species that can be found in various areas. Yet it is not impossible to get a list of varieties of species that could indicate there was a body in the soil as microbiomes seep through from decomposition. This study determines the viability of using soil microbial DNA as a method of locating clandestine graves by testing 6 different locations of a previous pig decomposition simulation. These two locations give two different scenarios that a body may be found either exposed to the sun in an open field or hidden under foliage such as a tree in the Sonoran Desert. The experiment will also determine more factors that could contribute to a correlation of microbiome specific groups associated with decomposition in soil such as firmicutes. The use of soil microbial DNA testing could open the doors to more interpretation of information to eventually be on par with the forensic use of biological DNA testing which could potentially supplement testimonies on assumed burial locations that occurs frequently in criminal cases of body relocation and reburial.
ContributorsMata Salinas, Jennifer (Author) / Marshall, Pamela (Thesis director) / Bolhofner , Katelyn (Committee member) / Wang, Yue (Committee member) / Barrett, The Honors College (Contributor) / School of Mathematical and Natural Sciences (Contributor) / School of Humanities, Arts, and Cultural Studies (Contributor)
Created2022-05