Last week I attended the Strategies and Techniques for Analyzing Microbial Population Structures (STAMPS) Course in Woods Hole, MA. This course was recommended to me by my advisor and a collaborator from the Garcia-Pichel Lab at ASU (Phoenix). The purpose of the course is to teach people who are new to coding the process of taking metagenomic data (raw sequences) and turning that into a publishable product like a figure for a paper. The course covered two different "metagenomic" strategies. The first is called marker gene sequencing. You've probably heard of the 16S rRNA gene, which is a gene for part of the ribosome in bacteria and archaea. The equivalent gene for eukaryotes (like humans) is the 18S rRNA gene. In this type of sequencing, you sequence only a small part of the 16S rRNA gene, but you can sequence ALL of those genes in your sample. The other metagenomic approach is called shot gun sequencing. My favorite analogy for shot gun sequencing was provided by Susan Holmes at the STAMPS course: imagine that the DNA in your sample is a library full of books. Then the scientist comes along and rips up all the books and their pages into millions of little shreds containing only partial pieces of sentences. The job of the analysis is to put some of those pieces back together so that you can read the book. You might miss pages, you might miss entire books. But you get some of the information back through this process. The course was 10 days long and was about 8 hours a day. We were taught by some of the most prestigious individuals in the field and their graduate students who are the ones writing the software that we will be using. It was really awesome to actually meet these people and to see how much they really cared about making their software useable and accessible: which means teaching it to us!
Some of the topics covered included: basic coding in R and Unix, data visualization, metagenomic workflows, basic statistics, phyloseq, heterogeneous data structures, using controls in your work, trimming, assembly, taxonomic annotation, phylogenetics, estimation of microbial abundance and diversity, multivariate analysis, strain-level analysis, functional analysis, and multi'omics. As you can see, we covered a lot and you can find more details about each of these here. Of course, you'll need a git hub account to be able to access it, so not quite open to all. For those reading this blog, I think the most important thing to convey is how difficult it is to figure out which microbes are present and in what quantity. You'd think this was really basic and you're probably saying, "haven't we figured this out?" The answer is no. There are so many reasons (statistical and instrumental) why we cannot measure the exact number of microbial organisms ("species" present in a sample...this is called diversity. For one, it is super difficult to say how related two microbes need to be in order to be called a single species. Do they need to match 100% of their genetic sequence? How about 97% similarity? My favorite part of STAMPS was meeting all of the other students. We were majority graduate students but there were also post-docs and faculty. People from other countries. Mostly, the students study human microbiomes, particularly the gut microbiome. But there were a few of us working on environmental microbiomes- birds, soils, corals, sediments, water. Many of us would visit the beach (2 minute walk from campus) daily and on our off day, many of us visited Martha's Vineyard for a bike ride around the island.
0 Comments
As part of the Interdisciplinary Quantitative Biology program this year, we were asked to work in teams of three on collaborative science projects. These teams were mentored by IQ Bio faculty and were designed as a Grand Challenge, where each team was given a single open-ended research question. The goal was to narrow down the question into a manageable project, work with stakeholders to develop meaningful solutions, and present collaborative solutions.
It was important to me that this team science component of graduate training remain in the curriculum, so I interviewed the students and faculty who participated and provided feedback to the administrators of the program. The following year, I repeated these interviews to assess the efficacy of the changes that were implemented based on feedback from year 1. I intend to continue these interviews over the course of my PhD and provide at least 4 years worth of feedback on various quantitative team science efforts at the graduate level via publication. These interviews are made possible each year by members of the first-year cohort who are trained to interview their peers and mentors. The IPCC, UN, and others have dedicated major resources toward understanding global land degradation and toward science-policy makers discussing ways of avoiding, reducing, and reversing land degradation. We are using document analysis and topic modeling to understand the alignment in science-policy recommendations across these reports to prioritize strategies and understand linkages between those recommendations and the Sustainable Development Goals.
|
AuthorSierra is a graduate student in the Barger Lab at CU Boulder studying microbial ecology for dryland restoration. Archives
August 2023
Categories |