27 innovations from Bar-Ilan University, available for licensing, co-investment, or spin-out through BIRAD.
Sol Efroni
The invention is a computational framework that predicts a subject's biological age and immunological state directly from the sequence composition of their T-cell receptor (TCR) β-chain repertoire. It combines two complementary feature representations - age-associated clonotype abundances and CDR3 amino-acid 3-mer frequencies - which are processed through an ensemble of gradient-boosted trees for feature selection and a deep multi-layer perceptron (MLP) neural network for regression. A novel signed-Wasserstein distance scoring method identifies individual TCR clonotypes statistically biased toward younger or older donors, creating an interpretable molecular 'aging vocabulary' that feeds the predictive model. The system achieves robust cross-cohort age prediction (MAE ≈ 6–7.5 years, R² ≈ 0.67–0.78) and reveals a reproducible ~1.5-year sex-related immunological offset, with the capacity to reconstruct a subject's history of viral exposures from the TCR repertoire alone.
Knisbacher Binyamin
The invention presents methods for customizing chronic lymphocytic leukemia (CLL) treatment through the use of a detailed biomarker panel and specific agents, based on the patient's biomarker profile. This method aims to match patients with treatments likely to be effective, according to various markers from distinct categories (e.g., gene expression, protein expression, mutations, clinical features, disease subtypes), aiming to improve treatment precision and outcomes in CLL management. The approach suggests a shift towards personalized treatment, leveraging molecular characteristics to inform therapeutic choices in oncology. The results are, at least in part, based on dynamic BH3 profiling experiments, which were used in a comprehensive drug screen that tested 42 different FDA approved drugs on samples taken from dozens of CLL patients.
Yissachar Nissan
The invention relates to the identification and use of a unique gut microbes configuration that inhibits tumor growth and enhances the efficacy of immune checkpoint inhibitors (ICIs) in metastatic melanoma patients. By utilizing specific gut bacterial strains with immunomodulatory properties - it is possible to predict clinical response prior to immunotherapy, enhance anti-tumor immune responses, overcome treatment resistance, and improve therapeutic outcomes in patients with resistant cancers. Specifically, the invention focuses on a defined consortium of bacterial strains identified through a series of experiments conducted in our laboratory (by analyzing intestinal responses to patients microbiota samples using our unique gut organ culture system). This microbial mix includes Barnesiella intestinihominis, Ruminococcus callidus, Ruminiclostridium siraeum, and additional strains predicted to induce pro-inflammatory immune responses that potentiate enhanced anti-tumor immunity. The invention further comprises a novel experimental pipeline for identifying beneficial gut bacterial strains from among the thousands found in patient microbiome samples, aimed at boosting immune system activity and improving the success of anti-cancer treatments.
Louzoun Yoram
The effect of microbes on their human host is often mediated through changes in metabolite concentrations. As such, multiple tools have been proposed to predict metabolitc profiles from microbial taxa frequencies, assuming a direct relation between the gut microbiome composition and blood metabolite concentrations. However, the microbiome-metabolite relation may depend on host demographics or condition. We show that the relation between microbiome and metabolites is best predicted at the log concentration level. We further develop LOCATE (Latent Of miCrobiome And meTabolites rElations), a machine learning (ML) tool based on latent representation which predicts the log normalized metabolites composition based on the log normalized microbiome composition. LOCATE has a higher overall accuracy than all current state-of-the-art predictors in both 16S rRNA gene and shotgun gene sequencing. The accuracy of LOCATE and all other predictors significantly decreases when predicting on one dataset and testing on a different dataset, or on a different condition in the same dataset, especially in 16S rRNA gene sequence based data. We propose an intermediate representation between the microbiome and the metabolite concentrations and show that this representation can be used to predict the host phenotype better than either the microbiome or the metabolome. This representation is strongly correlated with host demographics, including age, gender and diet and can be used to improve ML predictions of host phenotypes in comparison with either microbiome or metabolome using a large microbiome sample combined with a small number of metabolome samples (~ 50)
Shamay Meir
Distal cis-regulatory elements, such as enhancers and silencers, dictate tissue-specific complex transcriptional repertoire in an orientation- and position-independent manner. Herpesviruses show programmed latent and lytic gene expression based on the infected tissue and physiological cell state. In a recent study we systematically identified the enhancers within the Kaposi’s Sarcoma-associated Herpesvirus (KSHV) genome. Here, we present ENHAvir, an NLM-based tool that can successfully predict the enhancers in a viral genome. We used the DeBERTa v3 language model56 in our framework. DeBERTa v3 is an encoder-style language model, making it a suitable candidate for extracting representations that could be used for new tasks. ENHAvir successfully identified known and novel enhancer elements in the herpesviruses, namely, EBV, HSV1, HSV2, VZV, HCMV, HHV-6, HSV-7, and MHV68. ENHAvir learned the minute patterns of previously published KSHV enhancers and their adjacent sequences responsible for enhancer looping. The activity of the predicted enhancers was validated by cloning the ENHAvir predicted sequences on the EBV genome downstream to the luciferase gene in a reporter plasmid with a weak promoter and performing dual-luciferase reporter assays in EBV-infected and uninfected cell lines. Interestingly, ENHAvir also precisely identified enhancers with the human genome, and examples for Fos, Jun, DPPA3, and Myc genes are presented. The ability of ENHAvir to predict both viral and cellular enhancers, provides an additional layer to the complex gene regulation via viral enhancers but also points out the evolutionary conservation of enhancer micro-signatures between a virus and its host.
Cohen Haim
Aging is associated with detrimental changes in chromatin structure and gene expression, contributing to inflammation, metabolic decline and tissue dysfunction. SIRT6, a histone deacetylase, plays a key role in maintaining chromatin integrity and promoting longevity. Here, we characterized age-related changes in chromatin accessibility in the murine liver. We found that aging leads to increased chromatin accessibility. These changes were accompanied by upregulation of inflammation-related pathways and downregulation of metabolic pathways. Remarkably, SIRT6 overexpression reversed these changes, reducing inflammation and enhancing metabolic function. Notably, ETS family members were enriched in regions with increased accessibility during aging, while liver-enriched transcription factors (LETFs) were enriched in regions with reduced accessibility. H3K9ac and H3K56ac ChIP-seq analyses showed that H3K9ac, but not H3K56ac, is associated with increased accessibility during aging and that SIRT6 can reverse this effect. Furthermore, an viral system of AAV-mediated SIRT6 overexpression experiment in aged mice demonstrated that SIRT6 not only slows age-related chromatin changes but can also reverse them, rejuvenating chromatin accessibility to a youthful state. This highlights the potential of SIRT6 based therapy to rejuvenate aged tissues and mitigate age-related dysfunction.
Byk Gerardo
Retinitis pigmentosa (RP) is a leading cause of incurable genetic blindness. Currently retinal gene therapies are mostly based on adeno-associated viral (AAV) vectors that are: i) limited in cargo size, limiting the delivery of large genes (e.g. MYO7A, a common cause for Usher 1B syndromic RP); ii) induce immune responses that may damage the retina, lower gene expression and preclude repeated treatment; iii) injected subretinally, treating a limited area, and associated with risks of retinal detachment and chorioretinal atrophy. Transient transfection in primary human non-dividing cells remains inefficient due to poor nuclear uptake and transient gene expression, typically lasting only 24–96 hours. Conventional methods like lipofection and electroporation often cause cytotoxicity and immune activation, reducing cell viability and reproducibility. Improved delivery systems are needed to overcome these barriers and achieve safe, targeted, and sustained transgene expression. We developed biodegradable nanohydrogels (NHGs) that form complexes with DNA. The new NHGs and their DNA complexes are devoid of cell toxicity, which, together with their tuned sizes, makes them potential tools for gene delivery and foreign protein expression. We demonstrate that these NHGs can be utilized to deliver genes to express foreign proteins in non-dividing human iPSC-retinal pigment epithelium cells, which are notoriously difficult to transfect. Controlling for DNA:NHG ratio, we can control for timing, duration, and levels of gene expression in human cells. Long duration of gene expression is easily achieved with protein expressed for at least 30 days post NHG treatment. Moreover, we developed a novel suprachoroidal delivery method that resulted in injected material distribution covering 80% of the retina in rats, rabbits, and monkeys. Thus, this invention includes: (1) a novel system for retinal gene therapy delivery for patients with retinal/macular degeneration; (2) a novel platform for sustained gene expression in human/other hard-to-transfect non-dividing primary cells that may be used for studying gene expression, disease etiology, etc.
alon shahar
We present a method for multiplexed RNA and protein detection in intact, thick organoids. Organoid thickness represents a major barrier to in situ molecular analysis. Our approach extends Expansion Sequencing (ExSeq)—previously limited to tissue sections up to ~50 µm—to organoids ranging from 100 µm to 300 µm in diameter. Importantly, organoids exceeding 300–500 µm frequently develop necrotic cores due to restricted oxygen and reagent diffusion, making 300 µm a practical upper limit for intact, physiologically relevant analysis. A key innovation of this method is delayed hydrogel polymerization, which enables uniform diffusion of gel monomers throughout the tissue prior to crosslinking. Additional challenges, including limited surface area, enzyme penetration, and imaging depth, were addressed through automated pipetting, glass-slide embedding, and protocol optimizations (Table). By physically expanding whole brain organoids within a hydrogel matrix, this method enhances spatial resolution by up to ~10× and allows iterative rounds of antibody staining and in situ RNA sequencing in the same sample. While demonstrated in neurodevelopmental organoid models such as STXBP1 encephalopathy, this platform is broadly applicable to diverse organoid systems for 3D multimodal molecular profiling.
Yavits Leonid
We present SquiBNet, a deep learning framework for large-scale bacterial pathogen identification directly from raw Oxford Nanopore Technologies (ONT) electrical signals, termed squiggles. Conventional nanopore pipelines require computationally expensive basecalling and reference alignment before classification. These stages introduce latency and preclude real-time operation on resource-constrained edge platforms. SquiBNet bypasses both stages and classifies organisms directly in signal space. The architecture combines a one-dimensional ConvNeXt-Base backbone with four Transformer encoder blocks. The convolutional backbone captures local signal structure. The Transformer head models long-range dependencies across the squiggle. Across 256 clinically relevant bacterial species, SquiBNet attains 82.07% test accuracy. This represents a 51.5 percentage-point absolute gain, approximately 2.7 times, over the strongest prior-art baseline adapted to the same task. Training relies on a validated large-scale simulated dataset constructed with Badread, Squiggulator, and slow5tools, comprising 18.5 million signal windows. Simulated signals were validated against real nanopore data through signal-to-noise ratio analysis and visual comparison. On the NVIDIA Jetson Orin NX edge platform, SquiBNet achieves a single-sample latency of 41.5 ms and sustains 32.5 samples per second at batch size 32, with negligible preprocessing overhead. The framework supports concurrent real-time classification of multiple nanopore channels, enabling point-of-care deployment within Read Until adaptive sequencing.
Byk Gerardo
The invention describes the synthesis of new nanoparticles with monodispersed sizes from 20 to 500 nm. The nanoparticles complex nucleic acids and is able to transfer genes into mammalian cells to express a foreign protein. The singularity of the nanoparticles is that they are highly biocompatible. The complexes can be incubated with cells for undetermined time without toxicity. The expression of the foreign protein start at low level after 48h but arrives to high level after 1 week with no toxicity. The transfected cells can be passed several times without loss of protein expression, indicating a controlled release of the nucleic acids and their expression. The invention includes the use of DNA or RNA. Finaly, in vivo experiments demonstrated that the gene can be administered for example SC or IM and the protein expression can be detected after 1 month.
Cohen Haim
To explore the role of protein post-translational modifications on lifespan and healthspan , we developed the PHARAOH computational tool based on the 100-fold differences in longevity within the mammalian class. Analyzing acetylome/phosphorylome and proteome data across 107 mammalian species identified 482 and 695 significant longevity-associated acetylated lysine residues in mice and humans, respectively. In addition, we have recognized 2115 longevity associated p phosphorylations. In regards to acetylations, these sites include acetylated lysines in short-lived mammals that were replaced by permanent acetylation or deacetylation mimickers, glutamine or arginine, respectively, in long-lived mammals. Conversely, glutamine or arginine residues in short-lived mammals were replaced by reversibly acetylated lysine in long-lived mammals. For phosphorylations, site these sites include phosphorylated Serine (S), Tyrosine (Y) and threonine (T) or their replacement to aspartic acid (D) or glutamic acid (E)or Alanine (A) and Y to phenylalanine (Y to F). Pathway analyses of the acetylation sites highlighted the involvement of mitochondrial translation, cell cycle, fatty acid oxidation, transsulfuration, DNA repair and others in longevity. A validation assay showed that substitution of lysine 386 with arginine in mouse cystathionine beta synthase, to attain the human sequence, increased the pro-longevity activity of this enzyme. Likewise, replacing the human ubiquitin-specific peptidase 10 acetylated lysine 714 with arginine as in short-lived mammals, reduced its anti-neoplastic function. These findings provide a computational tool for identifying modifications that control longer healthy life and potential interventions to extend human healthspan.
Okun Eitan
Down Syndrome (DS), which results from a trisomy of chromosome 21 (Hsa21), is the most prevalent genetic cause of intellectual disability worldwide1-3. Some genes on the human chromosome 21 are highly correlated to Alzheimer's disease (AD) pathogenesis, including the amyloid precursor protein (APP), which is overexpressed in individuals with DS, causing early-onset AD (EOAD)-related neuropathology4-6. Furthermore, an alarming association was found between pregnancy with a DS-affected fetus and a 5-fold increased risk of mothers developing late-onset AD (LOAD) later in life, compared with pregnancies with fetuses affected by other forms of intellectual disability7,8. We suggest that fetomaternal transfer of APP and its proteolytic fragments occurs during pregnancy with a DS fetus and that these fetal factors infiltrate the maternal central nervous system, contributing to Aβ-seeding in the maternal brain, adversely affecting their cognitive abilities. By modeling DS-like pregnancies in mice, by mating of wild-type (WT) female mice for 4 consecutive pregnancies with mouse strains that overexpress the human amyloid precursor protein (hAPP) during early embryonic stages, we have observed fetal DNA, mRNA, proteins, and cells in the mothers' brains and periphery tissues following pregnancies, resulting in maternal cognitive decline. To delay the observed cognitive decline resulting from the transfer of neurotoxic peptides, we have generated a novel vaccine specific for the human neurotoxic N-terminal epitope of APP, the N-APP (APP108-113) fragment, which we observed to reside in the periphery of Aβ plaques in the brain of EOAD mouse brains and present in the brains of mothers to hAPP-expressing mouse fetuses. Vaccinating WT female mice with the N-APP vaccination prior to their exposure to hAPP-expressing fetuses improved short-term memory abilities in several behavioral assays and may therefore be helpful as a preventative treatment for maternal cognitive decline following DS pregnancies. In addition, vaccinating 5xFAD mice, which model EOAD, with the N-APP vaccine, improved cognitive measures in various cognitive tasks.
