Welcome to the Brent Lab

    The Brent lab studies the causes and consequences of non-genetic variation in cell signaling and downstream phenotype.  Our current work studies systems level mechanisms by which evolved and synthetic S. cerevisiae signaling systems can limit the effects of this variability.   We are building on this knowledge to construct chemically tunable, variation suppressed controllers of gene expression in yeast and in higher eukaryotes.  By developing and distributing such "expression clamped" controllers to researchers worldwide, we hope to speed biological discovery by enabling investigators to tightly regulate and study incompletely penetrant phenotypes and threshold phenotypes affected by small differences in protein dosage.

     We have recently begun an applied project that uses deep neural networks that help augmented reality systems provide guidance for researchers performing lab procedures.

     These projects support the lab mission of working to accelerate the pace of scientific discovery over the course of the 21st century.  To this end, lab members are encouraged to consider the anthropology of the contemporary (ref), here including the regulatory, economic, political and social frameworks within which the lab's research functions, and the ways that increases in biological knowledge and capability are impacting human affairs.  

     Brent is a Professor of Basic Sciences and an Adjunct Professor of Public Health Sciences.  He holds affiliate appointments in the Department of Genome Sciences and Department of Bioengineering at the University of Washington.

Ref (Paul Rabinow, http://anthropos-lab.net/about.html)

News

Current Job Openings:

1. We are currently in search of a Research Technician III-IV to lead work on variation suppressed controllers of mammalian gene expression.

2. We are currently in search of a computer science, bio-engineering, or electrical engineering Work-Study Student to acquire, annotate, and curate image data for neural networks.

 

24 July 2019
Deep learning in Las Cruces

On July 23 and 24, Bill Peria, Roger Brent and collaborator Laura Boucheron led an intensive two day workshop on Deep Learning for Image Analysis at NMSI.  The ~30 attendees included professors, postdocs, USDA scientists, a few graduate students, and a single undergraduate, from different science and engineering disciplines.  Tutorials in Jupyter notebooks, written by Boucheron, taught participants Python coding to carry out basic image processing to build, train, and use convolutional neural networks within the Keras framework.  Brent, Peria and Boucheron articulated what they considered to be key topics for near term research.

18 June 2019
Two wonderful recent undergraduates

The lab is enriched by the presence of Gabriella LaBazzo, a microbiology major, from Cody, Wyoming by way of Colorado State, and Karrington T. Ogans, a mathematics major, from Seattle having graduated from Gonzaga University.  LaBazzo brings impressive infectious disease lab and field experience to her current work in and mammalian cells on Well Tempered Controllers.  Ogans, who has been with us since February, is now working on deep neural networks and their possible use in epistemic support.

6 June 2019
O frabjous day! Single 5' intron sufficient for good reporter expression in worms

Scientific Reports has accepted for publication Crane et al., "In vivo measurements reveal a single 5’-intron is sufficient to increase protein expression level in Caenorhabditis elegans" for publication.  During the scrambling early years of the recombinant DNA era and the start of the commercial biotech industry, it became widely known that inclusion of introns (Kaufman and Sharp, 1982, 1983, Kaufman, 1985) (the so called adenovirus tripartite leader, Logan and Shenk (1984) and polyadenylation sequences led to greater expression of recombinant proteins.  The stimulation of gene expression by [even single] 5' introns was quickly found to operate in plants (Callis, Fromm, and Walbot, 1987).  In C. elegans, very early work on transgenes by Andrew Fire and coworkers (Okema et al. 1993) showed introns aided gene expression and Fire included a three-intron construct in the clasical "C. elegans vector kit".  But there was never a good reason to think that there was anything magical about three 5' introns, and the current work (BiorXiv 499459) shows that, as expected, there isn't.  Work was started here by Bryan Sands and Alexander Mendenhall and finished in collaboration with them lab at the University of Washington.

12 May 2019
Actually, 4th lab supercomputer said Hello World already

As befits the nature of the work (to be funded by the Defense Threat Reduction Agency) the advent was a little stealthy... but the lab's 4th "supercomputer", Ghostwheel, has been up and running for awhile.  Most of its work is calculations needed for computer vision.  The most important part of Ghostwheel's brain is an NVIDIA RTX 2080; making him even more powerful than the first three lab computers.

10 March 2019
William Lai is a visiting scientist

Lab is enriched by William Lai.  Lai is a senior and distinguished engineer, comes as a visiting scientist.  Most of his early and formative industrial experience was with Microsoft.  After more than a decade there, in 2008 he left Microsoft with a friend to co-found of 8ninths.com, an Augmented Reality agency in Seattle, where he handled engineering and technical development.  Lai sold 8ninths successfully in late 2018. Lai will work on technology development projects underway here.