Genomic Research Uses

#AskDrTim 🔆 Why Everyone Is Talking About “Omics” If artificial intelligence is transforming healthcare from the outside, omics is transforming it from the inside. Omics represents one of the biggest shifts in modern medicine. Instead of studying a single gene, protein or biomarker, it examines entire biological systems to understand how our bodies function, age and respond to disease. Think of your body as a symphony orchestra. * Genomics is the musical score, your DNA blueprint. * Epigenomics is the conductor, determining which genes are switched on or off. * Transcriptomics captures the music currently being played, revealing which genes are active. * Proteomics studies the musicians, the proteins carrying out the work. * Metabolomics measures the performance itself, reflecting your body’s real-time metabolic health. * Microbiomics explores the trillions of microorganisms that influence immunity, metabolism and even brain health. Each layer tells only part of the story. Together, they create a comprehensive picture of human biology. This is why multi-omics is rapidly becoming the foundation of precision medicine and longevity medicine. Rather than asking: “What disease does this patient have?” We are increasingly asking: “What biological changes are occurring long before disease appears?” This paradigm shift allows us to move from reactive medicine to predictive, preventive and personalised healthcare. Imagine identifying accelerated biological ageing before symptoms develop, tailoring nutrition to your metabolism, selecting medications based on your genes, or detecting disease years earlier through molecular signatures. That future is already beginning. As I continue my journey into longevity medicine, I am increasingly convinced that the future of healthcare will not be defined by treating disease alone, but by understanding the complex biological networks that govern health, resilience and healthy ageing. The physician of tomorrow will not rely on a single laboratory result. They will integrate multi-omics, artificial intelligence and clinical judgment to deliver truly personalised care. The future of medicine isn’t about knowing more facts. It’s about seeing the whole biological picture. #LongevityMedicine #PrecisionMedicine #Omics #AskDrTim

Triple-negative breast cancer (TNBC) is one of the most aggressive breast cancer subtypes. It lacks the three receptors (ER/PR/HER2) that enable targeted therapies in other forms of breast cancer and recurs early (often peaking ~3 years after diagnosis). Its genomic instability and immunogenic microenvironment make it a strong candidate for individualized immunotherapy. In a Phase 1 clinical trial led by Prof. Dr. med. Marcus Schmidt and investigators from Germany and Sweden, just published in Nature, we evaluated an individualized neoantigen mRNA vaccine approach in 14 patients with early-stage TNBC after surgery and (neo)adjuvant therapy. Each vaccine encoded up to 20 patient-specific neoantigens on two mRNA molecules, delivered intravenously via lipid nanoparticles to target dendritic cells. The results showed robust immune responses: • All patients in the clinical trial developed vaccine-induced T cell responses against multiple neoantigens. • Vaccine-induced CD8⁺ T cells reached frequencies commonly achieved with adoptive T cell therapies and persisted functionally for years without boosters – evolving into both "ready-to-act" cytotoxic effector cells and stem-like memory T cells. • 11 of 14 patients remained relapse-free for up to six years post-vaccination. Furthermore, the findings in three patients with relapses were instructive for potential future combination treatment strategies to overcome resistance – each revealing a distinct escape mechanism to be addressed: • Enhancing response magnitude: The patient with the weakest vaccine-induced response relapsed but achieved complete remission on subsequent anti–PD-1, suggesting a response threshold and supporting combination strategies. • Targeting antigen-presentation loss: One patient showed near-complete loss of MHC class I (likely via B2M downregulation), despite vaccine-induced T cells being present, highlighting the need to address HLA-loss escape (e.g., antibodies or strategies restoring recognition). • Comprehensive tumor sequencing: In another patient the relapse originated from a contralateral, genetically independent tumor not covered by the vaccine design, underscoring the importance of sequencing multiple lesions in hereditary settings. Overall, these results demonstrate feasibility and durable neoantigen-specific immunity in TNBC supporting personalized mRNA cancer vaccines as platform technology, while pointing to novel treatment strategies to overcome resistance – especially through informed treatment combinations. 𝐋𝐢𝐧𝐤 𝐭𝐨 𝐩𝐮𝐛𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧: https://lnkd.in/dk4fq6nA #CancerResearch #Oncology

Thread: Multi-omics sounds cool—until you actually try it. Here's are the nuances. 1/ You’ve got RNA-seq. Methylation. Proteomics. Time to “integrate” the data. But how? And why? Let’s break it down. 2/ Multi-omic integration sounds powerful. But it’s not magic. If you don’t ask the right question first, the answer won’t matter. 3/ Start here: Do you want shared programs across omics? Or unique signals from each modality? That choice decides your method. 4/ Unsupervised goal? Try MOFA2. Want to predict disease or treatment? DIABLO is your friend. Graph models? Great—if it performs better 5/ Real-life example: Chronic kidney disease study used both MOFA2 + DIABLO. Why? Different tools, complementary insights. Paper: https://lnkd.in/eZ_Fu83u Another New preprint for a different disease: https://lnkd.in/esXGmdqQ 6/ Here’s what makes multi-omics hard: Your matrix is incomplete. RNA-seq for 200 samples. Proteomics for 150. Methylation for 180. 7/ You can’t just “merge” them. Naive concatenation drowns real signal. Or worse—creates phantom clusters driven by batch noise. 8/ Each modality is different: scATAC-seq is sparse Proteomics is noisy RNA-seq has 20K+ features Methylation may only cover 50K regions and over 9 million CpG sites 9/ Good methods normalize each modality, learn weights, or regularize smartly. MOFA2, DIABLO, and weighted PCA all do this. 10/ Want to see how it fails? Check this post: https://lnkd.in/eMiCtVgW Spatial + gene expression integration went sideways without normalization. 11/ Math is nice. But biology matters more. If you can’t map back your result to a gene, CpG, or protein—what’s the point? 12/ These methods uncover correlations, not causes. Interpret carefully. Validate everything. 13/ Use known pathways. Run orthogonal experiments. Generalize across cohorts. Don’t trust the output blindly. 14/ Resources: Tools list: https://lnkd.in/eri4hGKR Tool review: https://lnkd.in/etcQfBm4 Overview: https://lnkd.in/esK4M-eG 15/ Key takeaways: Start with the question Pick tools based on your goal Normalize per modality Validate everything Biology > black boxes Multi-omics is messy. But it’s worth it—if you know what you’re doing. I hope you've found this post helpful. Follow me for more. Subscribe to my FREE newsletter chatomics to learn bioinformatics https://lnkd.in/erw83Svn

HER2-positive. HER2-directed therapy. Suddenly, HER2 seems to be everywhere. Conference halls. Journal headlines. Drug approvals. Oncology discussions. And understandably so. Some of the most transformative advances in cancer medicine over the last two decades have come from HER2-directed therapies. But I sometimes wonder, how many people actually know what HER2 is? HER2 is not a cancer. It is a protein receptor found on the surface of cells. The gene responsible for making this receptor is called ERBB2. Think of HER2 like an antenna. Its role is to receive signals that tell cells when to grow and divide. In healthy cells, this signalling is tightly controlled but in some cancers, that control is lost or deficient. HER2 biology in simple terms: • HER2 overexpression = too much HER2 protein on the cell surface • HER2 amplification = extra copies of the ERBB2 gene driving excess HER2 production • HER2 mutation = changes in the DNA sequence that can switch HER2 signalling on abnormally These are not the same thing and that distinction matters. When we say a tumour is HER2-positive, we usually mean: ✔ HER2 overexpressed (IHC 3+) and/or ✔ ERBB2 amplified (on ISH) Why is this important? Because HER2 signalling can drive cancer growth but remarkably, it can also become a therapeutic vulnerability. This is precision oncology in action. Historically, HER2+ breast cancer carried a poorer prognosis. Today? That story looks very different. HER2-directed therapies have transformed outcomes for many patients. However, HER2 is not just a breast cancer story. We now recognise HER2 biology across multiple tumour types, including: • Breast • Gastric and gastro-oesophageal • Lung • Colorectal • Endometrial • Biliary tract • Ovarian • Salivary gland tumours This is why HER2 testing matters because identifying HER2 is not simply describing a tumour. It may identify a treatable target. The HER2 story continues to evolve. A reminder that cancer is not just about where it starts. Increasingly, it is about what drives it. We have entered an era where tumour biology has become as important as tumour geography. One of the most fascinating shifts is that we have moved beyond the old binary of HER2-positive versus HER2-negative, so on my next post, I will be breaking down what is meant by the term "HER2 low". Repost, share, like, follow and comment if you have learned something new. #DrBookiesNuggets #HER2 #Oncology #PrecisionMedicine #TargetedTherapy #CancerResearch #BreastCancer #CancerEducation

🚀 Our perspective is out in Nature Portfolio! We present a roadmap for Multimodal Foundation Models (MFMs) — large AI models pretrained across multi-omics and multi-timepoint data — to serve as the computational backbone for building virtual cells. Read the full paper in Nature: https://lnkd.in/g3yzvA_f 🔎 Why MFMs? Biology is inherently multimodal, and molecular layers are deeply interconnected and context-specific. MFMs aims to integrate these layers to uncover shared biological principles that govern diverse cell states, offering a unified substrate for downstream inference. 🧠 What’s new? 💡 From hypothesis-driven to data-centric workflows: MFMs shift biology’s paradigm. Instead of crafting bespoke models for narrow tasks, we can now pretrain over massive datasets, distill foundational knowledge, and refine insights through lab-in-the-loop experimentation—where models guide experiments, and experiments update models. 🧬 Conditional gene regulation: MFMs go beyond static models. By training across multiple omics layers (e.g., chromatin accessibility, transcriptomics), they can learn context-specific gene functions and regulatory programs—key to understanding development and disease. 🧪 In silico perturbation: Biology’s combinatorial complexity is immense—thousands of genes, millions of interactions. MFMs provide a framework to simulate perturbations before wet-lab execution. Trained on CRISPR perturb-seq data, they can predict molecular responses across cell types, tissues, and time—enabling programmable biology at scale. ⚙️ What makes MFMs possible? Envisioned techniques include: - Unified tokenization from nucleotides to pathways - Hybrid attention across intra- and inter-modal interactions - Prompt-driven multitasking for temporal prediction, conditional generation, and modality translation - Human knowledge integration from curated databases and biomedical literature These design principles translate the architecture of foundation models into the molecular domain. ⚠️ What are the challenges? MFMs aren’t just about scale—they demand accessibility, reliability, and transparency. - Low-resource learning techniques (e.g., LoRA, adapters) are vital for democratizing training - Human-agnostic benchmarks are needed, as conventional labels may punish models that uncover novel biology - Uncertainty modeling is essential to mitigate hallucinations and increase scientific trust Interpretability and ethical stewardship must be foundational in this emerging ecosystem. Kudos to all co-authors for the collective effort and vision: Haotian Cui Alejandro Tejada Lapuerta #MariaBrbic Julio Saez Rodriguez Simona Cristea Hani Goodarzi Mo Lotfollahi Fabian Theis Let’s build the future of virtual cells together.

Understanding Neurodegenerative Diseases Through the Lens of Multiomics Our Review published in Annals of Neurology: https://lnkd.in/giW3rZ4T Neurodegenerative diseases such as Alzheimer’s, Parkinson’s, Lewy body dementia, and frontotemporal dementia are rising rapidly as populations age. Despite their clinical differences, these disorders share complex, multifactorial biology that has long limited our ability to diagnose early, predict progression, or identify effective therapeutic targets. Our new review synthesizes what the field has learned from genomics, transcriptomics, and proteomics—and how these technologies are reshaping our understanding of disease mechanisms. Multiomic studies have: • Identified dozens of genetic loci across AD, PD, DLB, and FTD, revealing pathways involving immunity, lysosomal biology, lipid metabolism, and synaptic function. • Highlighted the power of rare variant discovery and whole‑genome sequencing to uncover causal genes and modifiers of disease risk and resilience. • Mapped transcriptomic changes across brain regions, cell types, and peripheral fluids, exposing early immune activation, mitochondrial dysfunction, and RNA dysregulation. • Enabled large‑scale proteomic profiling in CSF, plasma, and brain tissue, uncovering disease‑specific signatures, biological subtypes, and promising biomarker panels. • Demonstrated how QTL mapping and integrative analyses can connect genetic variation to molecular mechanisms, nominating causal and druggable targets. The takeaway is clear: multiomics is no longer a future promise—it is actively transforming how we define, diagnose, and study neurodegenerative diseases. The next frontier will require harmonization across platforms, deeper ancestry diversity, and unbiased measurement of molecular modifications. But the path toward biologically grounded precision medicine is now visible.

🔥 Minimal Residual Disease Beyond the smoke: How we're using science to find what's left behind 🔥 For generations, cancer care has centered on the visible. We use surgery, radiation, and chemotherapy to fight the blazing inferno of a tumor until the flames are gone and the smoke clears. Yet, too often, we know that despite our best efforts, the fire can return, sometimes months or even years later. This is because we have long lacked the tools to see what is left behind. Think about it: once a forest fire is out, you can look across the charred landscape and see no active flames. From a distance, it looks like a job well done. Our traditional methods of surveillance, like CT scans, are often doing just that—looking for the big plumes of smoke that signal the fire has started again. But what if, beneath the ashes, a few microscopic embers are still glowing? These are the minimal residual disease (MRD) cells—the silent, unseen remnants of the tumor. They are a ticking clock, a source of potential new growth, and the greatest threat to a lasting cure. An objective look at prognosis: This is where a profound shift is occurring in oncology. We are no longer limited to just looking for a new blaze. New molecular tests, powered by incredible advances in genomics and artificial intelligence, are giving us the ability to find those nearly invisible embers. These tests, often called liquid biopsies, analyze a patient's blood for tiny fragments of cancer DNA. This provides us with an objective, quantifiable measure of a patient's risk of recurrence. This is more than just a lab result; it is a new level of clinical confidence that allows us to make better decisions for our patients.     •    A negative test is like searching the ashes with a thermal sensor and finding nothing. It gives us powerful reassurance that the embers have been extinguished. This objective data can be the basis for de-escalating therapy, potentially sparing a patient the toxicity of unnecessary treatments.     •    A positive test is like finding a glowing ember in the dark. It is a clear signal that microscopic disease remains. This information can be the catalyst for escalating therapy, guiding us to new or extended treatments to put out the last of the fire before it can spread again. The more published studies I read, the more convinced I am that we are underutilizing these tests. These are IMO SOC now for colorectal cancers, and strong evidence now emerging this year supporting a number of other tumors like breast, bladder, melanoma, other GI with similar data. This ability to see what was once unseeable changes the entire conversation around cancer prognosis. It moves us from broad, population-based risk assessments to a highly personalized approach, giving us the power to truly tailor our treatments and give each patient the best possible chance for a long, healthy life, free from the threat of recurrence #MinimalResidualDisease #MRD #MCED

🧬 Whole genome analysis identifies many more cancer patients who could likely benefit from DNA-repair-targeting drugs Standard testing for homologous recombination deficiency (HRD) focuses on BRCA1/2 mutations and a few genomic scars. But HRD—the DNA repair defect that makes tumors vulnerable to PARP inhibitors and platinum chemotherapy—can arise through many other mechanisms. Our new study in Communications Medicine (https://lnkd.in/ecMjtYdT) used whole genome sequencing (WGS) to find what current tests are missing. This work comes from our precision medicine initiative with NewYork-Presbyterian and Illumina (https://lnkd.in/eRsyjV7X), where we deployed clinical-grade WGS at scale. The HRD classifier—developed by Isabl, Inc.—scans the entire genome for mutation signatures and structural patterns that indicate broken DNA repair. It's one example of what becomes possible when you have the full genome rather than targeted panels—and when academia and industry collaborate on clinical-scale genomics. 📊 What we found across 580 tumors We analyzed paired tumor-normal WGS from 453 patients across cancer types. HRD was most common in breast cancer (21%), followed by pancreaticobiliary (20%), gynecological (17%), and prostate (9%). 24% of HRD-positive cases were BRCA1/2 wild-type. 🔍 Beyond BRCA WGS revealed HRD mechanisms that targeted panels miss: deleterious structural variants disrupting DNA repair genes, plus genome-wide mutation signatures that reflect the underlying repair defect. 💡 Better prediction of treatment response When we compared our WGS-based classifier to commercial tests like MyChoice CDx and FoundationOne, the WGS approach showed stronger correlation with actual response to PARP inhibitors and platinum therapy. I think this reflects the richer information available when you can see the entire genome. 🎯 The bigger picture Beyond this study, I think WGS makes increasing economic sense: it costs a few thousand dollars—a fraction of a single cycle of many cancer therapies. And WGS captures not just point mutations and indels, but also structural variants that are hard to detect from targeted panels and even exomes. With that richer data, detecting complex mutational signatures like HRD is just one application—the same genome enables liquid biopsy monitoring to detect recurrence earlier than imaging, and provides the blueprint for personalized cancer vaccines. Congratulations to co-first authors Majd Al Assaad, Kevin Hadi, and Max Levine, and to Juan Miguel Mosquera who led this work with our team at the Englander Institute. More on the study: https://lnkd.in/erTF7taU