Satya Prakash Dash

(nearly) Everything you need to know for antibody developability & property prediction, a guide At Tamarind Bio, we gathered best practices + the most effective (mostly open) in silico tools across all key risks: Aggregation propensity: why it kills programs; sequence + structure tactics (Aggrescan, TANGO/Waltz, Aggrescan3D; HIC-ML surrogates). Solubility: sequence and structure predictors to rank and design mutations (NetSolP, Protein-Sol, CamSol, SOLpro/SOLart; DeepSol/PON-Sol). Viscosity & self-interaction: charge/hydrophobic patch analysis (SCM, DeepSP), B22/kD surrogates; practical cut-offs for high-conc mAbs. Stability & chemical liabilities: ΔΔG and hotspot scanning (ThermoMPNN; TEMPRO for VHH); deamidation/isoAsp, oxidation, glyco motifs, free Cys (abYsis, MusiteDeep). Immunogenicity: HLA-II centric risk scoring and deimmunization workflow (TLimmuno2, NetMHCIIpan, IEDB; B-cell: DiscoTope, ElliPro, ABadapt/EpiPred). pI & charge profile: fast screens + patch asymmetry (IPC/ProtParam; TAP, SCM). Manufacturability & expression: predicting titers and downstream pain via solubility/patch proxies; chromatography behavior cues (HIC/IEC). We also detail best practices for multi-stage filtering, evaluating multiple parameters e.g. solubility, stability, aggregation, distributions vs. clinical stage mAbs, along with how to benchmark and validate these tools. Read the full guide: https://lnkd.in/gS7rx4JK ——— Tamarind Bio is leading the collection of bioinformatics and computational molecular design tools like AlphaFold, RFdiffusion, (Ig)BLAST, ProteinMPNN accessible via our scientist-friendly web interface, AI Copilot and programmatic API.

228

7 Kommentare

⚠️ From Protocol to Publication: Where Indian Research Fails We don’t have a shortage of brilliant minds. We don’t lack patients, data, or disease burden. What we do lack is a system that lets research flow smoothly from protocol to publication. So where do we stumble? 🔹 Protocol Development Often copied from global templates, rarely tailored to Indian realities. 👉 Solution: Involve clinicians, statisticians, and patients early to localize protocols. 🔹 Ethics Review Slow, inconsistent, sometimes bureaucratic. 👉 Solution: Capacity building for IECs and harmonization of review processes. 🔹 Site Operations Overburdened investigators, under-trained CRCs, and poor infrastructure. 👉 Solution: Invest in site staff training, digital tools, and long-term site development instead of one-off trial setups. 🔹 Data Quality Paper-based chaos, delayed entries, and missing source documents. 👉 Solution: Mandate e-source/eCRFs, adopt CDM standards, and encourage real-time monitoring. 🔹 Publication Gap Too many studies end in conference abstracts. Very few reach peer-reviewed journals. 👉 Solution: Incentivize publications, provide writing support, and build a national repository for trial outcomes. India has the numbers. India has the brains. But unless we fix the pipeline from protocol to publication, we’ll always be the “data providers” for the rest of the Globe, never the innovators leading the way. 💡 Question for you: What’s the one systemic issue you think India must fix first to truly shine in global research?

167

6 Kommentare

AI in healthcare is making headlines, but very few know how to make it serve real clinical systems. 🔥 Hundreds of thousands saw the teaser for this Decoded episode. Now the wait is over. In this new podcast, I talk with Parminder Bhatia, Chief AI Officer at GE HealthCare, to unpack what really drives adoption - not someday, but today. Trust, validation, real-world constraints, and why adoption is the KPI that matters. 🎧 Episode 4 is now live on Spotify, Apple, Youtube, and wherever you get your podcasts.

1.520

34 Kommentare

Hot off the press: Imagine being able to use the same kind of model as ChatGPT or Gemini to predict not the next word, but the next clinical event in a person's life. A groundbreaking new paper in Nature, "Learning the natural history of human disease with generative transformers," introduces Delphi-2M, a powerful new application of the GPT transformer architecture applied to the UK Biobank data to model the health course of volunteers, like narrative or story, and trains models the figure out how to predict the next part of the narrative of health for a different person. The model was trained on an incredible scale, using data from over 400,000 participants in the UK Biobank, a world-class population-scale research cohort, to train and then from another 100k of those participants to test and optimize hyper parameters. What’s truly exciting is how the model was then validated on a completely different population: 1.9 million Danish individuals. The implications of this work are immense. While the accuracy at the individual level may not be accurate enough for routine clinical decision making, aggregated at a population level, Delphi-2M could project the expected disease burden at local, regional, and national levels, helping healthcare systems prepare for future needs and guiding health prevention policies and strategies. And the opportunities to identify disease associations is huge—eg Figure 4B highlights some of the common conditions preceding a diagnosis of pancreatic cancer and preceding death for example. Some of these associations are known, and some are detectable only with AI. While the paper highlights the power of this approach, it also transparently discusses limitations and biases. The UK Biobank, like many research cohorts, has a "healthy volunteer bias," and its participants are not representative of the entire population, tending to be more affluent and educated. There's much opportunity to use these foundation models to achieve the goals of precision medicine, if we could get these data together. It’s also a great example about how important it is for us to understand the fundamentals of how these foundation models work so we can adapt them meaningfully to important use cases in clinical care. Kudos to the authors. Artem Shmatko Alexander Yung , Kumar Gaurav, Søren Brunak, Laust Hvas Mortensen, Ewan Birney, Tom Fitsgerald, Moritz Gerstung. EMBL European Molecular Biology Laboratory, University of Copenhagen (Københavns Universitet), Novo Nordisk Foundation foundatio  #AIinHealthcare #PrecisionMedicine #DigitalHealth https://lnkd.in/eGDRptp5

103

8 Kommentare