Larry Mays Seminars in Bioinformatics

Dr. Cory Bruower Presenting

UNC Charlotte’s Department of Bioinformatics and Genomics (BiG) hosts a weekly seminar series named after Dr. Lawrence Mays (Emeritus Faculty). The Larry Mays Seminars in Bioinformatics covers research within Bioinformatics, Genomics, and Computational Biology. Seminars are free and open to the public. Seminars take place on Fridays at 2:30- 3:30 PM. All seminars are hosted from the Bioinformatics building, room 105. Please contact Dr. Denis Jacob Machado or Dr. Laurel Yohe for questions or suggestions.

The seminars are also a one-credit graduate course (BINF-6600 and BINF-8600 combined). The instructor for the Fall 2023 semester is Dr. Jacob Machado. Dr. Yohe will be the instructor in the Spring semester of 2024. Students enrolled in BINF-6600 or BINF-8600 can see course details on CANVAS.

Schedule for the Fall semester of 2023:

  • 08/25: No seminars (we will host the Student Orientation and Five Minute Madness instead)
  • 09/01: No seminars (Labor Day weekend)
  • 09/08: Seminar 01—UNC Charlotte’s HPC student group will introduce a new training module for our research clusters
  • 09/15: Seminar 02—Dr. Ward Wheeler (American Museum of Natural History) will talk about multi-armed bandits, Thompson sampling, and machine learning in phylogenetic graph search, among other topics
  • 09/22: Seminar 03—Dr. Lucas Czeck (Carnegie Institution for Science) will talk about population genetic statistics for the next generation of pool sequencing.
  • 09/29: Seminar 04—Dr. Cathy Moore (Biosafety Officer at UNC Charlotte’s Division of Research) provides Biosafety Training for Graduate Students
  • 10/06: Seminar 05—Dr. Viviana Monje-Galvan (University at Buffalo) talks about an atomistic view of protein-lipid interactions at the cell membrane interface
  • 10/13: Seminar 06—Dr. Grace Wyngaard (James Madison University)
  • 10/20: Seminar 07—Dr. Brian Cleary (Boston University)
  • 10/27: No seminars (Football History vs. Florida Atlantic University at the Jerry Richardson Stadium)
  • 11/03: Seminar 08—Dr. Muhammad Iqbal Qureshi (Public Health Reference Laboratory and Khyber Medical University)
  • 11/10: No seminars (Veteran’s Day)
  • 11/17: Seminar 09—Dr. Carla Osthoff Ferreira de Barros (National Laboratory of Scientific Computing, Brazilian Ministry of Science, Technology and Innovation)
  • 11/24: No seminars (Thanksgiving Break)
  • 12/01: Seminar 10 (last one)—Dr. Joshua David Campbell (Boston University)

Note: This schedule is subject to change, and this website will be updated every Tuesday during the Fall semester of 2023. Zoom links will be shared via email (you may contact Dr. Jacob Machado for details).

Zoom link for Dr. Cathy Moore’s seminar: CLICK HERE (meeting ID: 995 4287 4731, passcode: safety).

PREVIOUS AND CURRENT EVENTS

I. September 8, 2023: UNC Charlotte’s HPC student group will introduce a new training module for our research clusters.

The Charlotte HPC Group gave the 9/8 seminar. This new student organization is focused on advancing high-performance computational skills for users of all levels across campus. They do this through various means, including in-person workshops, peer support, and online Canvas training. This day, they briefly showed the new module and told us about some of the other activities they have on the horizon. This seminar was informative for people who are new to the cluster or don’t even have an account and for more experienced users who want to expand their knowledge and find additional resources. The module is already available to the 49er community. Please use this form to request additional information.

II. September 15, 2023: Dr. Ward Wheeler talks about multi-armed bandits, Thompson sampling, and machine learning in phylogenetic graph search, among other topics.

Dr. Ward Wheeler visited in person on 9/15 coming from the Division of Invertebrate Zoology American Museum of Natural History, New York, NY, USA. In his talk, he talked about something cool called the multi-armed bandit problem in the context of phylogenetic graph searching. He uses a technique called Thompson sampling on a bunch of ‘search-bandits’ to help focus on search strategies that work best. This smart random sampling method turns out to be super effective in creating really good phylogenetic graphs, and it’s quicker than other random search methods. It’s like an unsupervised machine learning strategy that we can use on all sorts of phylogenetic data, even if we don’t know much about them beforehand.

III. September 22, 2023: Dr. Lucas Czeck talks about population genetic statistics for the next generation of pool sequencing.

In recent decades, so-called Evolve-and-Resequence (E&R) experiments have become a popular approach to survey rapid evolution of populations over multiple generations. These experiments allow us to measure shifts in the allele frequencies of a population in response to new or shifting environmental conditions, such as a changing climate.

Pool-sequencing of several individuals at once is a cost-effective and efficient tool to obtain reliable allele frequencies from a population of thousands to hundreds of thousands of individuals, and is often used in E&R experiments. However, specialized tools to efficiently analyze these data that take sampling noise stemming from the pool-sequencing approach into account were lacking. We developed two software tools to overcome statistical and bioinformatic challenges arising in this context.

First, we present grenepipe, a workflow from raw sequencing data of individuals or pooled populations to genotypes (variant calling) and population allele frequencies. The pipeline automates trimming, mapping, variant calling, and quality control, with a selection of popular software tools in each of these steps, and produces variant calls and frequency tables. While generally applicable to individual sample data, it offers specialized steps for pool-sequencing. Our software downloads all dependencies and runs all steps automatically using a single command line call, and parallelizes processing for computer cluster environments, allowing large datasets to be analyzed efficiently.

Second, to enable inferences of evolutionary signatures from frequency data, we created grenedalf, a C++ command line tool to compute population genetic statistics. It computes unbiased statistics of Fst, Pi, Tajima’s D with pool-sequencing data, far outperforming alternative tools, and fixing long-standing issues in the existing tools. Further it offers novel data exploration tools such as windowed allele frequency spectrum visualizations based on the allele frequencies, and built-in data filters and manipulations. Apart from its speed, it offers many convenience options, such as reading all standard file formats.

These tools are designed for scalability and ease-of-use with contemporary file formats, which we showcase using the GrENE-net.org project, a large-scale Evolve-and-Resequence experiment with Arabidopsis thaliana from across the world. We will present some preliminary results form that experiment, showcasing the use of the above tools, as well as demonstrating how E&R experiments can be used to investigate rapid evolution under differing environmental conditions such as caused by climate change.

Recommended publications:

  • Click here for the grenepipe paper.
  • Click here for the grenedalf preprint (submitted to Bioinformatics).

IV. September 29, 2023: Dr. Cathy Moore (Biosafety Officer at UNC Charlotte’s Division of Research) gives a special presentation.

Dr. Cathy Moore provided Biosafety Training for Graduate Students.

Dr. Moore is the University contact for:

  • Handling and working with biological agents on campus
  • Institutional Biosafety Committee protocol review and approval process
  • Risk assessment and risk mitigation for biological agents and materials
  • Use of select agents
  • Dual Use of Research Concern
  • Biosafety training courses and requirements
  • Microbiological and Biomedical facilities renovation and design

V. October 6, 2023: Dr. Viviana Monje-Galvan (University at Buffalo) talks about an atomistic view of protein-lipid interactions at the cell membrane interface.

  • Talk description: My research group uses all-atom molecular dynamics to study lipid-lipid and lipid-protein interactions. One of our goals is to determine the role of specific lipid species in disease mechanisms. We specialize in developing realistic membrane models that reproduce the physical environment of a given organism or cell organelle using key lipid species for that particular membrane. In this talk, I will present a protein-membrane study in the context of hepatitis C. Specifically, the molecular signature that results upon protein binding and the associated local membrane response when different lipid species are used to model the membrane. These results help us understand the lateral reorganization of lipids around proteins and the potential impacts of chronic hepatitis C and lipid redistribution, particularly near fat storages of the cell.
  • Short Bio: Dr Monje-Galvan is originally from Bolivia; she completed her undergraduate and graduate studies in Chemical Engineering at the University of Maryland-College Park. She completed her postdoctoral training at the University of Chicago, where she continued strengthening her background in computational biophysics. She started her lab at the University at Buffalo in January 2021, where she leads a diverse group of graduate and undergraduate students. She is actively involved with AIChE and the Biophysical Society, as well as her university, in efforts to promote the participation of women and minorities in science and engineering. She is a board member of LatinXinChE, an affinity group within MAC AIChE, AIChE CoMSEF, and is an active volunteer with the Society for Biophysicists of Latinamerica (SoBLA).