Das Lab

First public presentation of (Part 1 of) the #Kaggle 3D #RNA Folding challenge results -- see you Thursday!

In 2 weeks, 9 Kaggle teams have beat AlphaFold 3 score (0.372) in predicting hidden 3D structures of #RNA, including complexes with small molecules & proteins. What will it take to beat the best template oracle (0.554)? www.kaggle.com/c/stanford-r...

OLE RNA from Clostridium botulinum (pdb_00009lcr) is shown in pink on the left. ROOL RNAs can form cages built from 6 (right, top, pdb_00009j6y) or 8 (right, bottom, pdb_00009mds) subunits.

Large and intricate naturally occurring structures composed exclusively of RNA... ROOL!
More at Molecule of the Month: pdb101.rcsb.org/motm...

Inspired by 2025 Stanford #RNA 3D folding
@kaggle
challenge outcome, we’ve collected cryo-EM of new-to-Nature RNA folds, RNA complexes with small molecules and proteins, and assemblies up to 6000 nts - and launched PART 2! Deadline: March 2026. www.kaggle.com/c/stanford-r...

Template-based RNA structure prediction advanced through a blind code competition Automatically predicting RNA 3D structure from sequence remains an unsolved challenge in biology and biotechnology. Here, we describe a Kaggle code competition engaging over 1700 teams and 43 previous...

Preprint on the Stanford #RNA 3D folding Kaggle challenge is out. My scientific new year’s resolution is to brush up on template-based modeling. #gofai
www.biorxiv.org/content/10.6...

The Eterna puzzle interface displaying the three states of the RNA polymerase ribozyme using RibonanzaNet in state 1 and Eternafold in states 2 and 3.

A new 3-state Replicase challenge has been launched on Eterna and will only be open for two weeks before sequences are collected for experimental testing at the Das Lab. Will Eterna players be able to improve the catalytic activity of the RNA polymerase?

eternagame.org/labs/14183416

Wonderful collab with Fu and Zuchero labs, led by @vedtopkar.bsky.social in our lab. How do molecules make it out to the myelin sheath? For the most expressed protein in oligodendrocytes, an #RNA switch appears to be the answer. Meng-meng's posts explain it best!

Compact RNA sensors for increasingly complex functions of multiple inputs - Nature Chemistry Designing single molecules capable of complex sensing functions is challenging. Now, using crowdsourced RNA designs from the online game Eterna, compact single-molecule sensors have been demonstrated ...

@eternagame got single RNAs to compute functions like all logic gates, A>B, and A*B/C^2, relevant for tuberculosis detection! Now looking for partners to scale to complex applications. Collab w/ Greenleaf Khatri labs @christianchoe.bsky.social @stanfordmedicine.bsky.social at
doi.org/10.1038/s415...

R1260 CASP target, water and ions around RNA. Ground truth on left, blind prediction from SoutheRNA MD simulations on right.

#MD blind predictions of water/ions around Tetrahymena ribozyme #RNA matched cryo-EM data better than our baselines! Congrats #CASP16 + @rachael-kretsch.bsky.social + 23 brave groups. Still a ways to go — time for #DeepLearning of water? Community paper: doi.org/10.1002/prot...

CASP RNA SIG Welcome nucleic acid enthusiasts to the Special Interest Group (SIG) focused different aspects of nucleic acid 3D structure prediction, including but not limited to obtaining experimental data, buildi...

Our work presented today at the www.youtube.com/@CASPRNASIG online seminar! @bussigio.bsky.social will report on the #CASP16 solvation challenge and 2D ensemble reconstructions using #DMS data

Overlay of top blind model and cryo-EM structure for OLE RNA complex, only CASP16 RNA target showing strong improvements over prior templates

Assessment of blind 3D structure modeling of #RNA and nucleic acid complexes now out in the #CASP16 special issue, led by @RachaelKretsch. We hope this benchmark, paper, metrics, and expert baselines drive much-needed progress by #CASP17 (May 2026)! doi.org/10.1002/prot...

The results are in: top codes in Stanford #RNA 3D Folding @kaggle.com competition are competitive with CASP16-leading humans Vfold, beat AlphaFold 3. Top team’s trick was template-based modeling, not #DeepLearning. Congrats: john, odat, Eigen, + all 1706 participants! www.kaggle.com/competitions...

Meet @frameshifter.org, the new nonprofit program established to steward Eterna's future and advance public engagement in scientific research through games, collective and hybrid intelligence, and human centered design.

Community science presents an alternative discovery framework for precision ASO drugs. A community voting system facilitates the testing of top designs, followed by experimentation to evaluate and score how well player-designed drugs ameliorated molecular disease phenotypes bit.ly/44u5oTv

Highly collaborative efforts that involved 6 lab members and the labs of @yiliangding.bsky.social and @rdaslab.bsky.social. We uncovered a telomerase dimer and its potential role in telomere maintenance.