Bioluminescence technology offers highly sensitive and non-invasive imaging in living organisms without the need for external excitation. Naturally occurring luciferases, the enzymes responsible for catalyzing light emission, constrained the full potential of luminescence technology for the past several decades due to their poor protein folding, large size, ATP dependency, and low efficiency.Creation of the next generation of luciferases required breaking free of evolutionary constraints. This work describes the creation of novel bioluminescent enzymes that surpass qualities of native luciferase using AI-powered de novo protein design. These designer luciferase catalysts enable genetic labeling across molecular, cellular, and individual levels in a multiplexed manner, using the same underlying technology.This advancement showcases the design of efficient enzymes from scratch in which our de novo luciferases will enable researchers to study complex biological phenomena effectively.In the last three decades, the development of fluorescent protein families has brought a revolution in the way researchers study biological processes in living cells. However, the dependency on external excitation for FPs introduces inherent drawbacks, such as phototoxicity and autofluorescence background. These especially limit the applications for fluorescent proteins in vivo. Bioluminescence technologies, which rely on an enzyme-catalyzed chemiluminescent reaction of a chromophore substrate to emit photons without the need for external light sources, circumvent these limitations and offer several orders-of-magnitude-higher sensitivity than fluorescence for macro-scale imaging.Practically implementing luciferases as general molecular proges has not progressed as far as fluroescent proteins due to a number of factors. Firefly luciferase (FLuc) is used widely for in vivo imaging, but it is dim, large (61 kDa), and ATP dependent. Gaussia luciferase (GLuc) is brighter than FLuc, but has five disulfide bonds and therefore cannot be used intracellularly. It is also prone to misfolding. Engineered variants of Renilla luciferase (RLuc) and Oplophorus Luciferase (NLuc) are brighter and more stable, but they emit blue light and have poor substrate specificity and therefore are difficult to used in multiplexed applications. LuxSit luciferase (Monod Bio Inc.) is the first de novo designed luciferase and has superior folding fidelity and stability to natural luciferases, but more de novo luciferase species are necessary to meet the needs of researchers.  

Kainate receptors, also known as kainic acid receptors are ionotropic receptors that bind to and are responsive to glutamate in neurons. These were originally identified as being activated by the compund kainic acid, orignally isolated from algae. Postsynaptic kainate receptors are involved in excitatory neurotransmission while presynaptic kainate receptors are involved in inhibitory neurotransmission. Kainic acid is a potentially very useful compound but very difficult to synthesize. As a result, there are very few pharmacological tool compounds to study kainate receptors and none that are readily tunable to install labeling compounds. 

The Major Histocompatibility Complex (MHC), is a genomic region that expresses proteins involved in immune system functions and that are important for organ transplantation. In humans, this type of gene is referred to as the Human Leukocyte Antigen (HLA). The HLA region is haplotypic, with all of the region inherited from one parent. HLA is highly polymorphic within the human population, both in terms of protein structure as well as genomic variability.This high genomic diversity makes accurate genotyping difficult using methods such as short-read sequencing. That said, current long-read sequencing methods and analysis can yield incomplete and inaccurate results. 

Clustered regularly interspaced short palindromic repeats (CRISPR) screening is a cornerstone of functional genomics, enabling genome-wide knockout studies to identify genes involved in specific cellular processes or disease pathways. The success of CRISPR screens depends critically on the design of effective guide RNA (gRNA) libraries that maximize on-target activity while minimizing off-target effects. Current CRISPR screening lacks tools that can natively integrate next-generation sequencing (NGS) data for context-specific gRNA design, despite the wealth of genomic and transcriptomic information available from modern sequencing approaches. Traditional gRNA design tools have relied on static libraries with limited genome annotations and outdated scoring methods, lacking the flexibility to incorporate context-specific genomic information. Off-target effects are also a concern, with CRISPR-Cas9 systems tolerating up to three mismatches between single guide RNA (sgRNA) and genomic DNA, potentially leading to unintended mutations that could disrupt essential genes and compromise genomic integrity. Additionally, standard CRISPR library preparation methods can introduce bias through PCR amplification and cloning steps, resulting in non-uniform gRNA representation.

Chinese hamster ovary (CHO) cells are a family of immortalized cells derived from the epithelial cells of the ovary of the Chinese hamster. They are among the most widely used cell systems in cell biology. The cell line is frequently used in biological and medical research. It’s specifically used in the production of recombinant therapeutic proteins. Essentially, the CHO cell is used as a “factory cell,” meaning it can be programmed to produce therapeutic proteins, including vaccines and antibodies.Recombinant protein expression has been used in the development of biologics for many applications including therapeutics and vaccines.  However, one challenge in the development of biologics is the unintended proteolysis associated of expressed recombinant proteins in CHO cells. One example of a biologic that faces supoptimal expression in CHO cells is the HIV envelope glycoprotein, gp120, which is one of the main targets for neutralizing antibodies in HIV infection and a prime candidate for component of an HIV vaccine. When expressed in CHO cells, gp120 undergoes proteolytic clipping by a serine protease at an epitope recognized by neutralizing antibodies. Essentially, the cells produce an enzyme that cuts gp120 into pieces. This proteolysis alters gp120 to the point where it is unrecognizable by the immune system and renders it non-immunogenic. This issue appears frequently with CHO expression of envelope proteins from clade B HIV. Clade B is the most common genetic subtype present in the US and Europe. While this issue has been observed and attributed to CHO cells, it was previously unknown which enzyme was responsible. Without knowledge of the enzyme responsible, the solutions were based on modifying the protein sequence or adjusting the conditions of production, both of which are suboptimal. 

Scaled neural sensing has been pursued for decades. Physical limitations associated with electrical (electrode-based) field recordings hinder advances in both field of view and spatial resolution. Electrochromic plasmonics (electro-plasmonics) has emerged as a rapidly advancing field combining traditional electrochromic materials with plasmonic nanostructures, including recent demonstrations of electrochromic-loaded plasmonic nanoantennas for optical voltage sensing. Existing optical electrophysiology techniques face critical limitations including poor signal-to-noise ratios due to low photon counts from genetically encoded voltage indicators, which have small cross-sections and low quantum yields. Fluorescent voltage indicators suffer from photobleaching, phototoxicity, and require genetic modifications that limit their clinical applicability. Current electrochromic devices also struggle with limited cycling stability, slow switching times, and restricted color options, and conventional plasmonic sensors exhibit inherently low electric field sensitivity due to high electron densities of metals like gold and silver. Current approaches to electro-plasmonics lack stable, high-contrast optical modulators that can operate at sub-millisecond speeds while maintaining human biocompatibility.

WiFi-based indoor positioning has been a widely researched area for the past five years, with systems traditionally relying on signal telemetry data including Received Signal Strength Indicator (RSSI), Channel State Information (CSI), and Fine Timing Measurement (FTM). However, adoption in practice has remained limited due to environmental challenges including signal fading, multipath effects, and interference that significantly impact positioning accuracy. Existing machine learning approaches typically require extensive manual feature engineering, preprocessing steps like filtering and data scaling, and struggle with missing or incomplete telemetry data while lacking flexibility across heterogeneous environments. Furthermore, there is currently no unified model capable of handling variations in telemetry data formats from different WiFi device vendors, use-case requirements, and environmental conditions, forcing practitioners to develop separate models for each specific deployment scenario.

Pressure injuries (commonly called bedsores or pressure ulcers) represent one of the most persistent and costly challenges in healthcare, affecting over 2.5 million US patients and costing almost $27B in 2019. Hospital-acquired pressure injury events occur in about 3% in general populations and about 6% in intensive care units (ICUs). Current prevention strategies still rely on the Braden Scale risk assessment tool as the gold standard. Developed in the 80s, it is used to stratify patients into risk categories based on factors like sensory perception, moisture, mobility, and friction. The Braden score directly informs turning frequency as the standard of protocol. Unfortunately, medical staff adherence to turning protocols remains low at ~50% nationally, creating a gap between prescribed care and actual implementation. Technologies to help assess by sensing pressure injuries have limitations, including discontinuous monitoring requiring manual interpretation, and lack of objective mobility metrics. These fail to account for the complex interplay between pressure distribution, patient movement patterns, and individual risk factors. The Braden-scoring approach is particularly problematic as it does not account for the presence of existing pressure injuries or patient-specific factors, and has been shown to have inadequate validity for ICU patients. Additionally, current pressure mapping systems are typically large, expensive, and require specialized training, limiting their practical deployment in routine clinical care.