zac.swider at gmail github.com/zacswider Google Scholar Profile
I have an extensive background using advanced fluorescence imaging approaches to uncover some of the many mysteries in cell biology. I completed my Ph.D. in Bill Bement's lab at UW-Madison where I optimized high-speed imaging approaches to study a fascinating dance between the Rho GTPases and their effectors. I also established simple and extremely effective computational approaches to quantify otherwise overwhelmingly complex time-lapse datasets. Over time I have transitioned to becoming primarily a software developer, where I use my background in the wet lab and behind the microscope to inform appropriate processing and analytical decisions. I currently work at Elephas Biosciences where my team and I are developing advanced 3D imaging and analysis approaches for live human tumor fragments.
2022 | Ph.D. | University of Wisconsin, Madison
2016 | Analytical and Quantitative Light Microscopy
2015 | B.S. | University of California, Santa Cruz
- 10 years of hands-on experience using advanced fluorescence imaging modalities with cells and tissues. Approaches include light-sheet, traditional widefield/confocal, two-photon/TCSPC, STED, SIM, and iSIM.
- An equivalent amount of experience optimizing sample prep for advanced imaging applications. Examples include engineering fluorescent protein conjugates (FPCs) to reduce toxicity, optimizing dosage and expression profiles for FPCs and/or exogenous dyes, and optimizing fixation, staining, and clearing conditions for immunofluorescence.
- Deep understanding of image formation, resolving power, and limitations of different imaging modalities. For those that know, I am always dancing around the pyramid of frustration.
- Strong programming skills using Python to process and analyze biomedical images. Extensive experience using distributed computing to process out-of-memory datasets and GPUs to accelerate bottleneck processing steps.
- Practical experience training and deploying machine learning and deep learning models for image restoration, object detection, classification, and 2/3D cell and tissue segmentation using PyTorch and Tensorflow.
- Extensive experience with team software development, including fluency with version control (git), environment control (conda, venv), testing (pytest), and a commitment to writing clean and well-documented code.
- Some experience developing image processing applications using C# and the .NET framework.
- I currently manage a team of scientists at Elephas Biosciences focused on developing advanced imaging and analysis approaches for two-photon/TCSPC images of human tumor fragments.
- Co-managed a shared imaging facility at UW-Madison (consisting of a Bruker swept field confocal, a Prairie point scanning confocal, an Olympus FV1000 confocal, and a Flamingo light sheet system) to maximize system use, minimize the impact of routine maintenance, and enhance system features. Educated new users on system features, analysis techniques, and best practices.
- Mentored 4 undergraduates over the course of 4 years to learn modern molecular biology and imaging approaches as well as scientific reading and writing proficiency.
- All of the basic molecular biology stuff: DNA/protein preps, DNA/protein gels, traditional/reverse/quantitative PCR, ELISA, Western blots, bacterial cloning, various different flavors of DNA recombination (Gibson, In-fusion cloning, traditional, etc). You get the idea.
- More experience than I'd like using small glass needles to perform micro-surgeries on or to microinject compounds (RNA, proteins, drugs) into various difference types of embryos.
- Extensive practical experience using insect and mammalian cell culture for knockdown screens, transformations, and protein expression.
- Four years of experience with genetic engineering in Drosophila (forward/reverse genetics, CRISPR knockout/in, RNAi).
Elephas Biosciences | Image Data Scientist, Team Leader | 2022 - Current
- I work collaboratively with wet-lab researchers, commercial operations, and other software developers to develop and deliver image processing and analysis software to meet commercial and R&D needs.
- Developing infrastructure (tested and versioned Python libraries) for centralized processing of multi-photon FLIM images including phasor-based probe un-mixing, metabolic measurements, and cell/tissue segmentation.
- Established a centralized database and pipeline for human-annotated images to allow continuous fine tuning of deep learning models for label-free segmentation and classification of human tissue.
- Constructed and deployed a fully automated pipeline to stitch out-of-memory montages and derivative datasets, decreasing processing time by over 15x and increasing quality compared to the previous state of the art.
Relevant publications:
University of Wisconsin – Madison | Ph.D. Student | 2017 - 2022
- Developed new software and experimental approaches to quantify oscillatory dynamics of fluorescent probes in time-lapse datasets. This software is used daily by lab members with no programming experience and has enabling analysis of previously undetected phenotypes.
- Identified and quantified mechanical coupling between F-actin and microtubule cytoskeleton and formin-mediated F-actin assembly along astral microtubules PMID: 31091161.
- Tested hardware and software for a modular and highly portable light-sheet microscope, identifying hardware limitations and bug fixes for increased imaging speed and precision.
Relevant publications:
- Cell cycle and developmental control of cortical excitability in Xenopus laevis *MBoC Early Career Paper Award
- A versatile cortical pattern-forming circuit based on Rho, F-actin, Ect2, and RGA-3/4
- Rho and F-actin self-organize within an artificial cell cortex
- Cross-talk-dependent cortical patterning of Rho GTPases during cell repair
- Spindle–F-actin interactions in mitotic spindles in an intact vertebrate epithelium
NHLBI, NIH | Postbacc Researcher | 2015 - 2017
- Used RNAi screening, high speed live imaging, and quantitative particle tracking to identify Kinesin-1 as a candidate motor and Pericentrin-Like-Protein as a candidate cargo for centriole motility.
- Used live imaging, reverse genetics, and two-hybrid screening to identify direct functional protein interactions necessary for cell division in Drosophila neuroblasts.
- Used live super-resolution imaging to identify mechanisms regulating myosin filament assembly.
Relevant publications:
- Pericentrin interacts with Kinesin-1 to drive centriole motility
- Fascetto interacting protein ensures proper cytokinesis and ploidy
- Actin dynamics and competition for myosin monomer govern the sequential amplification of myosin filaments
University of California, Santa Cruz | Undergraduate Researcher | 2013 – 2015 Advisor:
- Used microscopy-based forward genetic screen in Drosophila embryos to suggest a role for Drosophila Sec1/Munc18 homolog during cytokinesis.
- Used temperature-sensitive alleles of Drosophila Sec1/Munc18 to identify Drosophila Sec1/Munc18 as a critical player throughout early and late stages of mitosis.
Relevant publications:
**Oregon Institute of Marine Biology | 2013 | Undergraduate Researcher
- Used live imaging and immunofluorescence to identify a putative role for the Micrura alaskensis homolog of Orthodenticle in promoting the differentiation of cells that make up the primary ciliated band.
Kita*, Swider*, et al. (2019) Spindle–F-actin interactions in mitotic spindles in an intact vertebrate epithelium. Mol Biol Cell. 30(14):1645-1654. *co-first authors
[Swider*, Ng*, et al. (2019) Fascetto interacting protein ensures proper cytokinesis and ploidy. Mol Biol Cell. 30(8):992-1007.](https://www.molbiolcell.org/doi/10.1091/mbc.E18-09-0573?url_ver=Z39.88-2003&rfr_id=ori:rid:crossref.org&rfr_dat=cr_pub%20%200pubmed) *co-first authors