Primary human intestinal models for DMPK/ADME and NAMs-based drug discovery. Built to replace Caco-2 and animal testing with human-relevant biology. RepliGut®
Your gut lining is bigger than you think—about the size of a tennis court 🧠➡️🎾. It’s a key player in nutrient absorption and defense. That’s why we created RepliGut—to model the real human gut, microvilli and all. Real science, real insights. #Biotech#GutHealth
@Organoids_MDPI@Uni_WUE Skipping past dissociated single cells and using an intact millimeter-scale vessel segment instead is the clever part. You get the whole stem cell niche instead of trying to rebuild it from scratch.
@smokinscientist Turns out 2D and 3D models disagree on drug targets for boring reasons: one reacts to growth format, the other to what's in the culture medium. Kind of satisfying to see it actually mapped out..
@BiosensingIns The spatial resolution catching cell-to-cell heterogeneity is the real value here. A bulk assay would've just given you one averaged number and hidden this entirely.
@FrontPharmacol The honesty about 'benchmark accuracy versus prospective validation' is what most drug-discovery AI coverage skips. That gap is the whole ballgame right now
@Organoids_MDPI@CAS__Science Silicon ions doing completely different jobs in gut cells versus bone cells is the kind of nuance that gets flattened in most 'ceramics for tissue engineering' hype. Good that they call it out directly.
@TheCureCircle Keeps the hour but not the minutes' is a great way to describe the limitation. Right cell types, wrong tempo, and that gap is exactly where organoid biology gets hard.
@BrunoSilva_MD Fatigue tracking with a stress-immune signature instead of CRP or disease activity is the more interesting finding here. It's decoupled from the thing everyone assumes drives it.
@NUFeinbergMed@NatureComms Epithelial cells getting treated as short-lived responders forever, and this flips that. A durable imprint from one metabolite signal is a bigger conceptual shift than the colitis-protection headline.
@PJOnline_News@MHRAgovuk Product consistency and biological variability being called out explicitly is the real challenge here. Living organisms don't behave like small molecules batch to batch.
@Biomol_MDPI Reported hit rates jumping from under 1% to 15%+ sounds great until you notice it's all happening in the same narrow slice of chemical space these models were trained on.
@ivasanthaNathan@nchembio Self-driving labs sound flashy, but the real test is whether the loop stays closed without a human re-steering it every few cycles.
@MPs_MDPI Treating drug-target interaction as a frequency-matching problem instead of a shape-matching problem is a genuinely different lever for the 'undruggable protein' problem.
@ilyassahinMD TNFα going from 'basically nobody gets hits here' to 12 working binders is either a genuinely big deal or a fluke of this one target. Curious which..
@DrSamuelBHume The gap closing almost entirely by FDA submission stage is the real story. All that variation between therapeutic areas happens way before the agency even sees it
@pschwllr@SchwallerGroup Experts couldn't tell its routes apart from published human syntheses. That's basically the Turing test for organic chemistry and it just happened quietly.
@GMFHx@NatMetabolism Putting synthesis, signaling, and translational potential in one place instead of three separate literatures is exactly the kind of paper that gets cited for a decade.