MISL Research Lab
Various Presenters (Allen School)
Research Talk
Thursday, December 7, 2023, 3:30 pm
Abstract
Chandler Peterson
Title: Boolean Search using Strand Displacement in DNA Data Storage
Abstract:
DNA has emerged as an ideal storage medium due to its high data density, durability, and enduring relevance. We introduce a scalable approach to integrating DNA strand displacement computation with high-density DNA data storage. We use DNA gates to perform computation and introduce a novel data storage format where each “data-gate” contains a file payload fragment and unique file address activated via strand displacement. To demonstrate our workflow, we perform random access through Boolean search in a small DNA database of eight encoded Haiku files. A programmable logic array with eight parallel AND gates map two file attribute inputs (e.g. poet: Bashō AND season: Autumn) to a unique address output for each file in the database. This enables algorithmic queries where file subsets can be selected based on attribute constraints. Experimental results indicate we can reliably read out single and multi-file subset queries and relative gate activation with next-generation sequencing.
Bio: Chandler Petersen is a 4th year PhD student advised by Prof. Georg Seelig. His research interests lie in DNA computing and molecular programming, and synthetic biology. His research focus is on scaling up DNA logic circuits using high-throughput synthesis and sequencing.
Cady Pearce
Title: Building fast, robust molecular circuits and amplifiers
Abstract:
Molecular computation is still in its early stages and a paramount concern is that these systems operate error-free with high probability in their designed contexts (e.g., interfacing with biological and chemical systems). We are working to create better computing components that are robust by design, can scale with the help of supporting software tools and novel molecular architectures, and operate orders of magnitude faster than previously demonstrated by leveraging ideas from material science. With these new ideas we also demonstrate a DNA-only signal amplifier capable of sensitive molecular detection within ten minutes.
Bio: Cady is a third year PhD student in the Thachuk lab. She is interested in wet laboratory automation and building molecular circuits to interface with small analytes in biological systems.
Carina Imburgia
Title: CRISPR-Cas9 for Similarity Search in DNA Data Storage Systems
Abstract:
Synthetic DNA is emerging as a promising medium for digital data storage, owing to its exceptional data density and longevity. Efficient DNA data storage necessitates selective information retrieval to reduce decoding time and costs. This work explores CRISPR-Cas9, a sequence-programmable nuclease, as a tool for content-based similarity search by using a trained encoder to map image features to Cas9 target DNA sequences based on the similarity of their semantic embeddings. We evaluated our model and architecture using simulated and laboratory Cas9-based retrieval experiments on an encoded image database and present our results showing the viability of the approach.
Bio: Carina is a 2nd year PhD student working with Jeff Nivala. She is interested in unconventional computing, hardware, and security.
Daphne Kontogiorgos-Heintz
Title: Modeling and decoding nanopore signals for single-molecule protein sequencing and barcoding
Abstract:
The application of nanopore sensor technology in protein sequencing holds significant promise for the detailed characterization of proteins at the single-molecule level. We introduce a theoretical framework designed to predict signals generated by proteins as they pass through a nanopore, based on their unique amino acid sequences. In parallel, we have crafted machine learning algorithms that can pinpoint individual amino acid mutations from variations in the ionic currents observed during protein translocation. While our long-term goal is the de novo sequencing of native proteins, our current research has successfully demonstrated precise sequencing of synthetic proteins, which has substantial implications for high-throughput assays and the field of synthetic biology.
Bio: Daphne is a second-year PhD student in MISL, advised by Jeff Nivala. Her research focuses on applying machine learning to biological problems. She completed her undergraduate studies at the University of Pennsylvania, where she studied Computer Science and Bioengineering.
Jason Hoffman
Title: FluoroPhone: Affordable Detection of Molecular Output
Abstract:
We present FluoroPhone, a system that leverages the ubiquity of complex sensors and compute power available on modern smartphones and minimally adapts them for molecular readout. Building upon ideas in previous studies that adapt smartphones for this purpose, this work aims to eschew unnecessary complexity in hardware setup. Towards this goal we have demonstrated that fluorometric and colorimetric readout of DNA Strand Displacement (DSD) reactions can be detected with a smartphone and minimal setup. We plan to further investigate these and other readout techniques based on the smartphone and develop the system towards a facile, inexpensive alternative for the detection of molecular output.
Bio: Jason is a 5th-year PhD student working on research in the MISL Lab and the Ubicomp Lab at UW. He is broadly interested in research that increases access to medical care, and has been focusing on smartphone-based molecular sensing and diagnostics.