Fanying Tang

Now online in @CD_AACR: Machine Learning Predicts Hepatocellular Carcinoma Risk from Routine Clinical Data: A Large Population-Based Multicentric Study - by @JClusmann, @jnkath, @C_V_Schneider, and colleagues @RWTH @tudresden_de

BREAKING: Claude can now research like a Stanford PhD student. Here are 9 insane Claude prompts that turn 40+ research papers into structured literature reviews, knowledge maps, and research gaps in minutes (Save this)

Thrilled to announce alphagenome-pytorch, an accurate, readable, and careful port of AlphaGenome's architecture and weights to PyTorch. Work with @gtcaa @m_kjellberg @chriswzou @tuxinming as part of the GenomicsxAI initiative between @anshulkundaje and @pkoo562 labs.

Integration of imaging-based and sequencing-based spatial omics mapping on the same tissue section via DBiTplus nature.com/articles/s4159…

Fragmentomic liquid biopsy enables early breast cancer detection, molecular subtyping and lymph node assessment nature.com/articles/s4146…

Mutations in certain genes are associated with worse survival outcomes among patients treated with Lu-PSMA; this suggests a role for more intensive surveillance to capture early progression. ow.ly/OXQP50YmVv6 #NuclearMedicine #ProstateCancer @AbbyPepinMD

Retrospective real-world clinicogenomic study in @Nature 5893 pts with mCRPC 37% HRR pathogenic variants. 13% BRCA1/2. 389 received olaparib: BRCA2: median OS 17.5 mo BRCA1: 8.1 mo HR 2.23 (P=0.008) BRCA1 clearly worse. Within BRCA2, subtype defines benefit ⚠️ BRCA2 loss: median OS 24.3 mo HR 0.42 (P<0.001) This is intra-BRCA functional heterogeneity ‼️ Not all BRCA alterations are biologically or therapeutically equivalent. rdcu.be/e6mxn @OncoAlert

Now online: A Circulating GPNMB-Based Multimodal Model Integrates Tumor-Immune Crosstalk to Predict Immunotherapy Response in Esophageal Cancer doi.org/10.1158/2159-8…

Changing training data alone can improve deep learning prediction of spatial transcriptomics gene expression from histology images by 38% (without any changes to model architecture). We've updated our preprint showing this with expanded results: biorxiv.org/content/10.110… 🧵👇 1/4

‼️Major step forward for multicancer blood testing. A new Nature Cancer study shows that layering fragment patterns on top of DNA methylation makes these blood tests significantly more accurate. At very high specificity, detection improved from 63% to 77%. #Lungcancer survival depends heavily on stage at diagnosis. Earlier & more accurate detection could be transformative. #LCSM nature.com/articles/s4301…

De-escalation, expansion and personalization of perioperative ICI👇@NatRevClinOncol ctDNA clearance & pCR are enabling risk and response-adapted treatment duration. Distinguising prognostic from predictive biomarkers remains critical. @OncoAlert @urotoday nature.com/articles/s4157…

𝗔 𝗧𝗶𝗺𝗲 𝗦𝗲𝗿𝗶𝗲𝘀 𝗙𝗼𝘂𝗻𝗱𝗮𝘁𝗶𝗼𝗻 𝗠𝗼𝗱𝗲𝗹 𝗕𝘆 𝗚𝗼𝗼𝗴𝗹𝗲 This has been pre-trained on a time series corpus of 100 billion data points, & shows impressive performance on various benchmarks from diverse domains. 𝗧𝗶𝗺𝗲𝘀𝗙𝗠 𝗚𝗶𝘁𝗵𝘂𝗯 𝗽𝗮𝗴𝗲: github.com/google-researc… 𝗟𝗲𝗮𝗿𝗻 𝗠𝗟 𝗮𝗻𝗱 𝗙𝗼𝗿𝗲𝗰𝗮𝘀𝘁𝗶𝗻𝗴: leanpub.com/pycaretbook/

The Nature Genetics paper is out today (!!) This one is honestly a dream for me. We review and add our perspective on how mathematical & computational modeling can help quantify and predict the evolutionary dynamics of clonal hematopoiesis (CH). nature.com/articles/s4158…

ESMO 2026 reshapes advanced prostate cancer care From de novo mCSPC to molecularly defined mCRPC, treatment is now earlier, intensified, and genotype-driven. 🧬 mCSPC 🟢 Low volume → ADT + ARPI is standard 🟣 High volume → Fit patients: ADT + docetaxel + ARPI (triplet) → Unfit for docetaxel: ADT + ARPI ⚠️ ADT alone is obsolete except in frail patients ❌ Zoledronic acid / denosumab NOT recommended in mCSPC 🧠 Relapsed mCSPC • Low volume: ADT + ARPI • High volume: ADT + ARPI ± docetaxel (selected) 🧬 mCRPC without known mutations • Prior ADT only: ARPI or docetaxel • Post-ARPI: docetaxel or ¹⁷⁷Lu-PSMA-617 • Post-docetaxel + ARPI: ¹⁷⁷Lu-PSMA-617 or cabazitaxel 🧬 mCRPC with genetic alterations • BRCA: PARP inhibitor ± ARPI > chemotherapy • PALB2 / CDK12: selective PARPi use • dMMR: PD-1 / PD-L1 inhibitor early • Aggressive variant / NEPC: platinum-based chemotherapy 🦴 Bone health ✅ All mCRPC with bone mets must receive BPA ❌ Not for mCSPC 💡 Takeaway Earlier intensification. Mandatory genomics. Radioligand therapy mainstream. 🔖 Save this for daily OPD decisions 📖 Full paper in comment ⬇️ #OncoTwitter #MedTwitter #ProstateCancer #GUOncology @OncoAlert @myesmo @esmo_open @asco @Annals_Oncology

Circulating tumor DNA as a biomarker in early phase clinical trials dlvr.it/TQxjNB

If CS224n teaches transformers, CS25 explores where they break and how they evolve. Highly recommended if you already understand the basics. They have an enitire playlist.

Clonal evolution and transcriptional plasticity shape metastatic dissemination routes in prostate cancer | Nature Communications nature.com/articles/s4146…

CellRank protocol now in @NatureProtocols! 🎉 @PhilippWeiler's detailed guide shows how to: - Integrate velocity, pseudotime & time points - Infer cell fate probabilities - Analyze lineage priming at scale Great complement to our CellRank 2 paper! 📖 nature.com/articles/s4159…

A study proposes combining genetic association data with Perturb-seq to infer causal links from genes to traits through regulatory mechanisms, offering a genome-scale framework to bridge genetic signals and biological function. nature.com/articles/s4158…

Great lightening talks by Beltran Lab members Balaji and Jake presenting their latest work! #AACRProstate26 @DanaFarber_GU