TURAKHIA LAB, UCSD
Prof. Yatish Turakhia
Welcome to Turakhia Lab!
I am an Assistant Professor in the Department of Electrical and Computer Engineering (ECE) at the University of California San Diego (UCSD) with a joint affiliation with the Department of Computer Science and Engineering (CSE), the Department of Bioengineering, and the Bioinformatics and Systems Biology (BISB) graduate program at UCSD. My lab is also affiliated with the Center of Machine-Integrated Computing and Security (MICS), the Center of Microbiome Innovation (CMI), and the Institute for Genomic Medicine (IGM) at UCSD.
My recent CV
{add other icons}2/10/2022 - RIVET manuscript published in
Latest News!2/15/2022 - Sumit Walia passes his PhD preliminary exam. Congratulations, Sumit!
2/10/2022 - RIVET manuscript published in Bioinformatics. Congratulations, Kyle!
x/xx/2022 - Yatish named MIT Technology Review Innovator Under 35
x/xx/2022 - Internship acceptance
x/xx/2021 - C&C committee
x/xx/2021 - SERC Scholar
x/xx/2021 - EECS CDEI Taskforce
x/xx/2021 - TA-ship for 6.034 (Artificial Intelligence)
x/xx/2021 - Online Learning facilitator for MIT
College of Computing course
Research Mission
We are an interdisciplinary group of researchers working on a broad range of problems at the cutting-edge intersection of computer engineering and bioinformatics. We work on problems that have enormous potential in biological and medical applications but where computational costs and speed impose a barrier.
Our research mission is to develop algorithms and domain- specific hardware accelerators that enable faster and cheaper progress in biology and medicine.
Our work is generously funded by XX, YY, and ZZ.
Sponsors & Collaborators
Ongoing Research
Pangenomics | to study genetic variation
Low-cost and high-throughput sequencing have enabled extensive genome sequencing, leading to the emergence of Pangenomics, an area that studies genetic variation within a species using multiple reference genomes. This research is set to impact numerous fields but requires a shift from traditional single-reference genome approaches.
Our team is developing new data structures, algorithms, and hardware acceleration to facilitate comprehensive and faster pangenomic analysis.Phylogenetics | to study evolutionary relationships
Phylogenetics studies the evolutionary relationships among biological sequences, organisms, or species, with applications in evolutionary biology, outbreak analysis,and cancer progression. Tackling the computational challenges posed by large phylogenies and the surge in genomic data needs developing new automated phylogenetic methods.
In collaboration with Prof. Siavash Mirarab's Lab at UCSD, our team is developing fully automated, scalable, hardware-software co-designed solutions for efficiently constructing phylogenetic trees.
Hardware Acceleration | of computational genomics
Genomic data is one of the fastest-growing data types on the planet and is far outpacing Moore’s law. From personalized medicine to species conservation, genomic data has far-reaching applications, but computational costs are posing ever greater challenges to exploit the full potential of this data.
We are developing novel algorithms and hardware (GPU/FPGA/ASIC) acceleration approaches to speed computational genomics tasks, such as genome assembly, read alignment, and whole-genome alignments, by orders of magnitude.Outbreak Analysis | real-time phylogenetics for pandemic analysis
With over 16 million (and counting) whole SARS-CoV-2 genomes sequences already, the COVID-19 pathogen is the most sequenced pathogen in history. Phylogenetic analysis using these genomes has played a vital role in tracking the virus evolution and transmission, but is posing major computational challenges.
Our lab is working closely with Prof. Corbett-Detig’s lab at UC Santa Cruz in maintaining and refining UShER (the largest tool used by CDC?) - a comprehensive. phylogenetic tree, consisting of all available SARS-CoV-2 genomes.
Wastewater-based Epidemiology | for tracking pathogen evolution and transmission
Genomic data is one of the fastest-growing data types on the planet and is far outpacing Moore’s law. From personalized medicine to species conservation, genomic data has far-reaching applications, but computational costs are posing ever greater challenges to exploit the full potential of this data.
We are developing novel algorithms and hardware (GPU/FPGA/ASIC) acceleration approaches to speed computational genomics tasks, such as genome assembly, read alignment, and whole-genome alignments, by orders of magnitude.Prior Research
Can Physical Tools that Adapt their Shape based on a Learner's Performance Help in Motor Skill Training?
Dishita Turakhia, Yini Qi, Lotta-Gili Blumberg, Andrew Wong, Stefanie Mueller
[To appear] In Proceedings of TEI ’21.
Project Page | PDF (Pre-print)
Thirteen Ways of Looking (SMArchS Thesis)
Can creative thinking be computational?
In this inquiry, I investigate what aspects of human intelligence enables us to be a creative society. I apply the computational framework of Shape Grammar developed by Prof. George Stiny at MIT to creative thinking for machines proposing to consider thoughts as shapes, ideas as spatial relations, and creative thinking as computing with shape grammar.
Project Page | PDF
sketchCAD
Using machine learning for architectural applications
Architects spend 60% of their design time in drafting their conceptual sketches into CAD drawings. In this project, we use machine learning to automate the mundane parts of drafting for increased efficiency.
Project Page | PDF
Minimalision
A device to minimize visual distractions
In today’s age of ubiquitous advertising, we are constantly bombarded by bits of information vying for our attention. The “noise” of our environment has reached a fever pitch for almost all of our senses. We believe that combining this filtering process with technology in a human-machine symbiotic intervention can help augment our ability to focus - and, in turn, help us kick the bad habit of constantly diverting our attention to technology.
Project Page | PDF
Spatial Cognition in Architecture
Overview of Scientific Aspects of Spatial Cognition and Its Application in Architectural Design
[Publication] In Proceedings of ANFA
In today’s age of ubiquitous advertising, we are constantly bombarded by bits of information vying for our attention. The “noise” of our environment has reached a fever pitch for almost all of our senses. We believe that combining this filtering process with technology in a human-machine symbiotic intervention can help augment our ability to focus - and, in turn, help us kick the bad habit of constantly diverting our attention to technology.
Project Page | PDF | Talk
Hand in Glove
Can the artistry of glass-blowing be augmented using sensor-gloves and robotic arms?
In this project we designed tangible interface that can be multisensory in nature - glows that can see, feel and make. Much of glass blowing artists’ knowledge is implicit in nature. By making multisensory glove, the aim of this project is convert this expert knowledge from being tacit to tangible. We designed hand gloves that function as an intermediary interface between glassblower’s palms and glass material. The gloves can detect subtle nuances in palm gestures, pressure and temperature changes while providing constant feedback to and from robotic arms that make the glass artifacts.
Project Page | PDF
Dynamic Tensegrity Systems
How can we use computation to design dynamic habitable spaces that respond to changing user needs?
[Publication] In Proceedings of CAADRIA
Tensegrity structures exhibit an exceptionally high strength-to-weight ratio and possess the unique property of being stable in zero-gravity spaces. I used Evolutionary Algorithmic to thoroughly investigate a set of arbitrary Tensegrity Structures which are tedious to design using traditional methods and determine new irregular and architecturally optimal forms.