Cover photo
Howard Salis
Works at The Pennsylvania State University
Lives in State College, PA
162 followers|75,483 views


Howard Salis

Shared publicly  - 
Gerd Moe-Behrens originally shared:
Leukippos Institute

The Bacterial Nanorecorder: Engineering E. coli to Function as a Chemical Recording Device
"Synthetic biology is an emerging branch of molecular biology that uses synthetic genetic constructs to create man-made cells or organisms that are capable of performing novel and/or useful applications. Using a synthetic chemically sensitive genetic toggle switch to activate appropriate fluorescent protein indicators (GFP, RFP) and a cell division inhibitor (minC), we have created a novel E. coli strain that can be used as a highly specific, yet simple and inexpensive chemical recording device. This biological “nanorecorder” can be used to determine both the type and the time at which a brief chemical exposure event has occurred. In particular, we show that the short–term exposure (15–30 min) of cells harboring this synthetic genetic circuit to small molecule signals (anhydrotetracycline or IPTG) triggered long-term and uniform cell elongation, with cell length being directly proportional to the time elapsed following a brief chemical exposure. This work demonstrates that facile modification of an existing genetic toggle switch can be exploited to generate a robust, biologically-based “nanorecorder” that could potentially be adapted to detect, respond and record a wide range of chemical stimuli that may vary over time and space."
Add a comment...

Howard Salis

Shared publicly  - 
The Ribosome Binding Site Calculator's "Design RBS with Constraints" mode @ always felt a bit slow. Sometimes, an RBS solution would only be found after 24+ hours of optimization. Other times, a solution would never be returned. The particular algorithm being employed is sensitive to the initial RBS sequence -- if you choose a sequence very far away from a good solution, it will take many hours to finish. If you choose a pretty good initial condition, the optimization will finish in minutes to hours.

I re-examined the RBS initialization code and saw quite a few areas for improvement. The new code does a better job handling multiple sequence constraints and mRNA folding constraints when generating initial RBS sequences. It appears to work faster -- there were about 10 jobs that been previously crunching away for days and they now returned solutions within an hour.

Faster optimization = less clicking 'My Results' over and over again. Yay!
Rob Egbert's profile photoHoward Salis's profile photo
Yes, indeed! Also, when entering the RBS Constraints, it helps to add as many Ns as you can clone in.

The following RBS Constraint might not be satisfiable with a high translation rate:

But adding more Ns will allow the optimization to find a good solution:

Note: if you don't need any constraints (ie. all Ns), then you can use the 'Design RBS' mode instead. It will generate solutions a bit faster as it can insert or delete nucleotides in the RBS, when necessary.
Add a comment...

Howard Salis

Shared publicly  - 
AVAILABLE IMMEDIATELY: Postdoctoral Positions in Synthetic Biology

Description: The Salis Lab at Penn State University is recruiting postdoctoral scholars to build a molecular security system for recombinant DNA. We seek highly motivated and productive candidates with a Ph.D. degree in the engineering or biological sciences as well as experience in molecular cloning and microbiology. Strong analytical and quantitative skills are required.

Applications should be a single PDF file containing:
1. A cover letter describing your research and career interests
2. Your CV
3. Any preprints of publications not yet available on your CV
4. Your answers to the DNA riddles posed at

Please have your most recent adviser send a letter of recommendation directly to me.

Applications should be sent to salis AT psu DOT edu using the subject line 'Postdoctoral application'.
Howard Salis's profile photo
Wow, it looks like I've stumped a lot of people with these DNA riddles. Does this say something about our ability to reverse engineer synthetic systems? (Hmm, there are certainly biosecurity implications here ...)
Add a comment...

Howard Salis

Shared publicly  - 
Based on recent CPU time statistics, the latest improvements to the DNA Compiler @ have greatly decreased the optimization algorithms' time-to-solution. It now takes much less time to design synthetic ribosome binding sites, RBS libraries, and bacterial operons with complex sequence constraints and with diverse protein coding sequences.

We recently released v1.1 of the RBS Calculator's free energy model, described at The free energy model more accurately predicts the translation initiation rate of mRNAs with weak 16S rRNA binding sites (Shine-Dalgarno sequences) and very strong secondary structures in the protein coding sequence (e.g. in GFP).

A v1.2 is also in the works. Perfection is a never-ending journey.

Thanks to everyone for helping to beta-test the algorithms.
Gerd Moe-Behrens's profile photo
Thanks for sharing and providing this nice software!
Add a comment...

Howard Salis

Shared publicly  - 
New version of the DNA Compiler Websuite @!

-- We've added a new design mode to the RBS Library Calculator that helps you design MAGE oligos for genome editing. Now you can optimize Ribosome Binding Site libraries to systematically vary the expression level of any protein in the E. coli genome, across a 100,000-fold range! You pick the protein, the minimum level, the maximum level, and the number of variants in your library. The optimization algorithm does the rest.

--- We've made several improvements to the "Design Ribosome Binding Site" and "Design Ribosome Binding Site with Constraints" mode to decrease their optimization times.

--- We've added a "Maximize Translation Rate" goal for both constrained and unconstrained design of ribosome binding sites.

--- We've added a "Job Status" indicator in the Results so you know when your job is on the queue, currently being run, or has been completed.

Please let me know if there's any issues. Comments, feedback welcome.
Gerd Moe-Behrens's profile photoHoward Salis's profile photo
The good thing about Web 2.0 interfaces: everyone beta-tests, whether they want to or not!

The bad thing about Web 2.0 interfaces: the beta-tester doesn't necessarily know when they hit a bug.

I give you the v1.2 known bug list:

1. in Design RBS with Constraints mode, if your initial condition for an RBS sequence has a different length than your RBS Constraint, your job will not run. Solution: throw an input validation error onto the web interface.

Temporary fix: please count your nucleotides.

(Bug list & fixes updated periodically.)
Add a comment...

Howard Salis

Shared publicly  - 
We have an amazing President.

House Republications are looking like anarchists now more than ever. Politics aside, who would vote for a Congressman that has so threatened our Nation's stability and rule of law?? Not I. is the official web site for the White House and President Barack Obama, the 44th President of the United States. This site is a source for information about the President, White House ...
Heto Ah Hi's profile photo
Mind blowingly unbelievable. Thanks for the share.
Add a comment...
Assistant Professor at Penn State University

Advancing the engineering science of Synthetic Biology!
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
State College, PA
New Jersey - New Jersey - Minnesota - California
Assistant Professor
  • The Pennsylvania State University
    Assistant Professor, 2010 - present
Basic Information