Evaluation of Cellulose-Matrix DNA Extraction and DadApp for Non-Invasive Paternity Analysis and Learning in Wild Baboons

Abstract

Since the 1990s, non-invasive genetic sampling has transformed research in genetics, evolutionary biology and ecological dynamics in natural animal populations. However, existing approaches are labour- and time-intensive relative to their limited data output, necessitating further optimization. Developing new laboratory methods, analytical tools and software can achieve this goal. This study evaluated the effectiveness of cellulose-matrix based nucleic acid extraction, developed a Shiny-based web application for non-invasive genotyping and paternity analysis in wild baboons and tested the suitability of the application as an educational tool in both molecular and computational aspects of parentage analysis. Faecal samples were collected from thirteen known individuals in one baboon troop from the Amboseli ecosystem in Kenya. Faecal samples were processed using two variant protocols of the cellulose extraction method (single Whatman and four Whatman) and a modified column-based extraction method as a positive control. While the cellulose extraction method failed to amplify DNA for paternity analysis using microsatellite primers, it showed potential for DNA extraction for bacterial identification using the 16S rRNA gene. Additionally, a Shiny web application named ‘DadApp’ https://kiragu-mwaura.shinyapps.io/dadapp/ was developed for paternity analysis using R programming language. Tested with validated non-invasive data from the ABRP, DadApp achieved a 100% success rate in correctly assigning paternity in confirmed triad relationships. In every case, the top log-likelihood ratio (LOD) scoring candidate father was consistent with the paternity assignment in the ABRP pedigree, which had been established through independent verification methods. Further, DadApp as an active educational tool, significantly improved participants’ knowledge of paternity concepts, with participants’ mean scores improving from 70.54 (range 45-90) on pre-test to 97.86 (range 80-100) on post-test. The difference was significant based on a paired Wilcoxon Signed-Rank test (p = 3.75 x 10-6), indicating enhanced learning outcomes. Additionally, participants’ interest and confidence in paternity inference were significantly enhanced after a practical session with DadApp (p-values = 9.485 x 10⁻⁵and 8.103 x 10⁻⁴, respectively). This research highlights the potential, despite limited success, of the cellulose extraction method in non-invasive genotyping and demonstrates the effectiveness of DadApp for paternity assignment and education.