Designing drugs around the proteins they must spare

Marc C. Deller, D.Phil.

h 2557 papers, ranked by citation19912026
Career
1991–2026
FGFR
2022
CalR-Ab
2023
CDK2
2024
KRAS
2025
JAK1
2026
Structure first
Anti-targets
Measured contacts
Pocket motifs
Property trade-offs
Speed
Twenty years
Sheet01/25Paperall fiveCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/aboutsitemdeller.com/publications

From Peanuts to Proteins

Enlarged print of an X-ray diffraction photograph taken at Leeds in July 1958
Leeds, July 1958. A lock of Mozart’s hair, given by his widow to Vincent Novello in 1829, Brotherton Library.
What anyone seesA grey blur with a hole in the middle.
αβ
What Astbury sawSpacings. Hair is a chain, coiled; pull it and the coil opens.
What it turned intoEvery structure after it. Same question, better cameras.

The press got the other half of it: the same man spun groundnut protein into a fibre and wore the coat to the BBC. A structure you can read is a material you can make.

William Astbury in the Textile Physics laboratory, University of Leeds, 1928Daily Mail clipping: His coat's made of pea-nuts, post-war fashionCartoon of a scientist in an overcoat made largely of pea-nutsCover of The Man in the Monkeynut Coat by Kersten T. Hall, Oxford University Press
Where it started
Sheet02/25PapercareerCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/aboutsitemdeller.com/structures

Twenty years from cytokine receptors to kinase drugs.

  1. Leeds 1991–1995 University of LeedsBSc, first classBiochemistry, molecular biology and biophysics; industrial year at Boots/Knoll.
  2. Oxford 1995–1999 University of OxfordD.Phil.Cytokines and their receptors; the first structure of human oncostatin M (1EVS).
  3. Yale 1999–2001 Yale School of MedicinePostdoctoral fellowJAK/STAT pathway proteins: into signal transduction and kinase biology.
  4. Pfizer 2001–2006 Agouron/PfizerSenior Scientist to Principal Research ScientistSBDD for oncology and immunology; IND support for cMET, ERK, VEGFR and HIV targets.
  5. Scripps 2006–2015 Joint Center for Structural GenomicsSenior Scientist to Target ManagerHigh-throughput structural genomics: 70+ novel structures, 40+ papers.
  6. Stanford 2015–2018 ChEM-H, Macromolecular Structure Knowledge CenterFounding DirectorBuilt a lab from scratch: protein production, structures, SSRL access for 70 users.
  7. Incyte 2018–2024 IncyteAssociate Director to Research FellowBuilt the $8M structural biology platform; 100+ co-crystal structures, 6+ INDs.
  8. Elora 2025– Elora TherapeuticsCo-founder and CSOEngineering PETase enzymes to clear microplastics from the body.
  9. DeepCovalent 2025– DeepCovalent Inc.Scientific Technical AdviserAI-driven small-molecule design: docking, co-folding, FEP and MD validation.
400+protein structures
57publications (Scopus)
2,800citations
25h-index
7+patents
6+IND filings
20years
Career
Sheet03/25Paper·Cpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/aboutsitemdeller.com/structures

Twenty years up and down one pathway.

NUCLEUSOUTINPETasemicroplasticOSMFGFGLP-1GLP1REnvE2CD4CD81JAK1JAK2TYK2STAT3STAT1STAT5βFGFR KDCDK2cyclin EDYRK1PKSSTINGGrb2ShcSosKRASRAFERKNANOG
gp130 · OSMR
Env · E2 HIV · Topo
mutCALR
FGFR1-4
PETase
GLP1R
JAK · STAT
ChEM-H
Grb2 · Sos
CDK2
KRAS · ERK
NANOG
Pathway
Sheet04/25PapercareerCpd·PDB4RBO
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/aboutsitemdeller.com/publications

I built the platforms and services that made campaigns faster.

Incyte platform, before and after

Crystallisation roboticsAutomated crystallisation and imaging: the front end of 100+ co-crystal structures at Incyte.
X-ray crystallography400+ structures over twenty years, with synchrotron work at SSRL and beyond.
Glacios cryo-EMCryo-EM brought into the Incyte platform for complexes that would not crystallise.
NMR · MS · HDXSolution methods for binding modes and mechanism of action when a crystal is not enough.
BiophysicsSPR, ITC and DSF: the binding and stability numbers behind the structures.
Protein productionConstruct design and expression in E. coli, insect, mammalian and cell-free systems.
CADD and AIDocking, FEP, MD, AlphaFold, Boltz and ProteinMPNN, used where they speed a decision.
Cloud and HPCAWS, SLURM and LIMS pipelines that move data from beamline to decision.
Team leadership20+ scientists mentored; 10+ CRO and vendor partnerships managed.
Vibe codingAI-assisted apps for science: 28 projects and 17 live web apps, this deck among them.
$8MIncyte platform
62unique targets
16antibody structures
307co-authors
10programmes
70Stanford users
17web apps
Career
Sheet05/25PaperIncyte · Scripps · StanfordCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/aboutsitemdeller.com/structures

A gene in. A molecule out. Nothing invented.

1
pET3a+4.6 kbKRAS1–169G12D6xHisTEVT7AmpR
The gene, put in a vector1–169
2
leaksIPTGBL21(DE3)
Bacteria told to build it37 °C
3
Ni-NTAwaste10 mM250 mMflow-throughKRAS
Caught by its histidine tag250 mM
4
sealedreservoirproteinwell200 nL + 200 nLvapour diffusion20 °C5 days
Two hundred nanolitres a drop200 nL
5
0.2 mmX-raysnylon loopω spindle
One crystal, in a loop0.2 mm
6
1.36 Åedgelunes, one turnno phases
The pattern it threwreciprocal
7
Static render of this view
The density that came out1.36 Å
8
The molecule built into it48 atoms
How a structure is made
Sheet06/25PaperKRAS G12DCpd5aPDB9E5D
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/krassitemdeller.com/structures

A mutation that writes its own target on the cell.

1234567891011121314structure solvedpatent deadline
DEEQRTRRMMRTKMRMRRMRRthe new C-terminus21 residues, nine of them arginineOUTINTPO-RJAK2JAK2CALR wild typenothing happensone frameshiftCALR*CALR*JAK2JAK2PPSTAT5PCALR mutantthe receptor is held onanti-mutCALRthe receptor is normal. the ligand is the disease.
Calreticulina chaperone, mostly in the ER
A frameshift at the tailone new C-terminus
Myeloproliferative neoplasmswhere the mutation is found
Why an antibodya neoepitope, not a pocket
first-in-classanti-mutant CALR antibody
>$500Mprojected annual sales
mutCALR 1 / 2
Sheet07/25PapermutCALRCpdFabPDBWO 2023/107994
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/calrsitemdeller.com/patents

Six contacts, and five of them are arginines.

Static render of this view
Arg7to Gly26 of the heavy chain2.8 Å
Arg11to Glu31, heavy chain2.5 Å
Arg17to Asp106, heavy chain3.0 Å
Arg20to Glu52 of the light chain2.4 Å
Arg20to Asp103, heavy chain3.2 Å
Arg21to Asp98 of the light chain3.0 Å
Eleven arginines, six readtake-home
mutCALR 2 / 2
Sheet08/25PapermutCALRCpdFvPDBWO 2023/107994
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/calrsitemdeller.com/patents

A drug is defined by what it must not hit.

FGFR3≠FGFR1 Spare FGFR1: no hyperphosphataemia
CDK2≠CDK1 Spare CDK1: an essential cell-cycle kinase
KRAS G12D≠KRAS WT Spare wild type: it is in every cell
JAK1≠JAK2 Spare JAK2: keep red cells and platelets
Selectivity
Sheet09/25Paperall fourCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/publicationssitemdeller.com/structures

Hit FGFR1 and phosphate rises; miss the mutants and cancer returns.

J. Med. Chem. 2022, 65, 15433-15442Discovery of Potent and Selective Inhibitors of Wild-Type and Gatekeeper Mutant Fibroblast Growth Factor Receptor (FGFR) 2/3Shvartsbart et al.10.1021/acs.jmedchem.2c01366

Upregulation of the fibroblast growth factor receptor (FGFR) signaling pathway has been implicated in multiple cancer types, including cholangiocarcinoma and bladder cancer. Consequently, small molecule inhibition of FGFR has emerged as a promising therapy for patients suffering from these diseases. First-generation pan-FGFR inhibitors, while highly effective, suffer from several drawbacks. These include treatment-related hyperphosphatemia and significant loss of potency for the mutant kinases. Herein, we present the discovery and optimization of novel FGFR2/3 inhibitors that largely maintain potency for the common gatekeeper mutants and have excellent selectivity over FGFR1. A combination of meticulous structure-activity relationship (SAR) analysis, structure-based drug design, and medicinal chemistry rationale ultimately led to compound 29, a potent and selective FGFR2/3 inhibitor with excellent in vitro absorption, distribution, metabolism, excretion (ADME), and pharmacokinetics in rat. A pharmacodynamic study of a closely related compound established that maximum inhibition of downstream ERK phosphorylation could be achieved with no significant effect on serum phosphate levels relative to vehicle.

OUTINFGFFGFFGFR1FGFR3sparehitproliferation in bladder and bile ductFGFR3 vs FGFR1
Static render of this view

A hydrogen bond drawn on a model, then measured in the crystal.

HingeAla567 backbone N-H3.0 Å
Gatekeeper Val564Val555 in FGFR3: the V555M/L resistance site
Active conformationKLIFS: DFG in, αC in
αC helixKLIFS: αC in, the active arrangement
Designed H-bond(S)-THF O to the Asn571 side chain3.0 Å
Water bridgePLIP's own category, drawn dotted
Hydrophobic contactsfaint and dotted, behind the polar picture
The bond you can design is not the bond that winstake-home
bioisostereIsopropyl to cyclic etherFGFR3 54 → 1.1 nM
permeabilityN-isopropyl pyrazoleCaco-2 0.2 → 1.5 (10⁻⁶ cm/s)
metabolismN-CD3 closes CYP3A4CYP3A4 9 → >25 µM
oral PKCompound 29 in ratF 82%
FGFR 1 / 2
Sheet10/25PaperFGFR3Cpd30PDB8E1X
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/fgfrpapermdeller.com/fgfrdoi

FGFR1 falls away while the gatekeeper mutants stay inhibited.

Static render of this view

FGFR3 vs FGFR1 IC50, every compound (nM)

Cell potency along the series (BaF3, nM)

0.5 nMFGFR3 IC50
82%rat oral F

Four edits built the potency; the cells followed.

FGFR3methoxy+15×
FGFR3cyclic ether+49×
FGFR3(S)-THF+2.7×
KITF + Me+75×
V555M 6.1 nMV555L 2.7 nMcompound 29, gatekeeper mutants
FGFR 2 / 2
Sheet11/25PaperFGFR3Cpd29PDB8E1X
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/fgfrpapermdeller.com/fgfrdoi

The screening hit was a better JAK2 inhibitor than CDK2.

J. Med. Chem. 2024, 67, 3112-3126Discovery of (4-Pyrazolyl)-2-aminopyrimidines as Potent and Selective Inhibitors of Cyclin-Dependent Kinase 2Hummel et al.10.1021/acs.jmedchem.3c02287

CDK2 is a critical regulator of the cell cycle. For a variety of human cancers, the dysregulation of CDK2/cyclin E1 can lead to tumor growth and proliferation. Historically, early efforts to develop CDK2 inhibitors with clinical applications proved unsuccessful due to challenges in achieving selectivity over off-target CDK isoforms with associated toxicity. In this report, we describe the discovery of (4-pyrazolyl)-2-aminopyrimidines as a potent class of CDK2 inhibitors that display selectivity over CDKs 1, 4, 6, 7, and 9. SAR studies led to the identification of compound 17, a kinase selective and highly potent CDK2 inhibitor (IC50 = 0.29 nM). The evaluation of 17 in CCNE1-amplified mouse models shows the pharmacodynamic inhibition of CDK2, measured by reduced Rb phosphorylation, and antitumor activity.

cyclin E1CDK2RbS phase, and the cell dividesfive other CDKs sit on the same cycleCDK1CDK4/6CDK7/9CDK2 vs CDK1
Static render of this view

Sulfonamide oxygens reach Asp86; the aminopyrimidine holds the hinge.

Hinge Leu83two H-bonds from the aminopyrimidine2.7 Å · 3.3 Å
Gatekeeper Phe80CF3 packs here; DFG in, αC out
DFG motifKLIFS: DFG in
αC helixKLIFS: αC out
Sulfonamide to Asp86backbone N-H3.0 Å
Lys89 and Gln85Lys89 direct; Gln85 through a water3.4 Å · 3.0 Å
Water bridgeGln85, reached only through a water
Hydrophobic contactsfaint and dotted, behind the polar picture
One swap changed the target, one atom bought the margintake-home
CDK2sp3 amine+38×
CDK1CF3+2×
SGFdiol+280×
CDK1cyclopropyl+1.5×
CDK2 1 / 3
Sheet12/25PaperCDK2Cpd17PDB8UV0
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/cdk2papermdeller.com/cdk2doi

One trifluoromethyl at the gatekeeper and CDK1 falls away.

Static render of this view

CDK2 vs CDK1 IC50, every compound (nM)

Solubility against selectivity, every compound

0.29 nMCDK2 IC50
220×over CDK1

Every fix traded solubility against selectivity; 17 won both.

CDK2Compound 1607 nM
CDK2Compound 90.42 nM
CDK1Compound 17220×
CDK2 2 / 3
Sheet13/25PaperCDK2Cpd17PDB8UV0
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/cdk2papermdeller.com/cdk2doi

The conformation was a calculation until the crystal agreed.

J. Med. Chem. 2026Discovery of INCB127443: A Potent and Orally Available Inhibitor of Cyclin-Dependent Kinase 2Mukai et al.10.1021/acs.jmedchem.6c01506

CDK2 is a critical regulator of cell-cycle progression and an attractive therapeutic target for cancers characterized by dysregulated cyclin E1 signaling. Despite this biological rationale, CDK2 has historically been challenging to target due to the difficulty of achieving selectivity over related CDK isoforms, particularly CDK1, where off-target inhibition may limit the therapeutic window. Building on our previously reported (4-pyrazolyl)-2-aminopyrimidine series, we pursued a multiparameter optimization strategy aimed at improving selectivity for CDK2 over CDK1 while balancing pharmacokinetic and physicochemical properties. Computationally guided replacement of the pyrazole with an imidazole led to a (4-imidazolyl)-2-aminopyrimidine series and culminated in the discovery of INCB127443, a potent and orally bioavailable CDK2 inhibitor.

One well, not twopyrazoleimidazoleenergy13GK: 174°0180360dihedral angle φ
Static render of this view

Selectivity over CDK1, five matched pairs

0.99 nMCDK2 IC50
316×over CDK1

Three hours on CDK2, fifty-three seconds on CDK1.

conformationImidazole for pyrazole221 → 568×
permeabilityA chlorocyanophenyl on the imidazoleCaco-2 1 → 9.5 ×10⁻⁶ cm/s
solubilityBenzylamine, then azetidineFaSSIF 76 → 895 µg/mL
candidateINCB127443monkey F 64%
residence time 192 minon CDK1 0.88 minINCB125496, TR-FRET probe competition
CDK2 3 / 3
Sheet14/25PaperCDK2Cpd30PDB13GK
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismpapermdeller.com/cdk2bdoi

Aspartate is not cysteine: no warhead, so shape must do it.

J. Med. Chem. 2025, 68, 1924-1939Discovery of INCB159020, an Orally Bioavailable KRAS G12D InhibitorYe et al.10.1021/acs.jmedchem.4c02662

The inhibition of mutant KRAS proteins has emerged as a promising approach for treating KRAS-driven cancers, as evidenced by the clinical success of KRAS G12C inhibitors. KRAS G12D, the most common mutant, promises significant expansion of the addressable patient population; however, the reduced nucleophilicity of aspartate compared to cysteine poses significant challenges in balancing sufficient potency with ADME properties to support oral exposure. Herein, we describe the discovery of KRAS G12D inhibitor 23 (INCB159020), which achieves oral exposure in nonhuman primate (NHP). Starting from a weakly potent hit, structure-based drug design was utilized to drive significant potency. Focus on molecular rigidity and balanced polarity then allowed for successful optimization of properties required for oral exposure.

OUTINKRAS WTKRAS G12DRAFERKin every cell37 per cent of pancreatic tumoursKRAS G12D vs KRAS WT
Static render of this view

A rigid cage holds the amine where Tyr96 can reach it.

Water to Gly10P-loop, through one water3.1 Å
Switch Iresidues 30 to 38, the effector region
Beside switch IIreversible, no warhead
Asp12: backbone onlywater to the N-H; side chain untouched3.0 Å
Gly13 and Gln61the catalytic machinery, untouched
Amine to Tyr96direct H-bond from the bicyclic amine2.8 Å
Water bridgetwo of them, drawn dotted
Hydrophobic contactsfaint and dotted, behind the polar picture
Shape has to do what a warhead doestake-home
chemotypeA rigid amine, found by screeningKRAS G12D 671 nM
potencyA linker reaching towards Asp1250×
dead endCompound 10 bottomed out the assaySPR <1 pM
candidateINCB159020, a thousandfold weakerNHP F 42%
KRAS 1 / 3
Sheet15/25PaperKRAS G12DCpd3PDB9E5F
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/kraspapermdeller.com/krasdoi

It grips the backbone beside Asp12, never the side chain.

Static render of this view

Whole-blood potency along the series (nM)

Permeability along the series (Caco-2)

22 nMG12D SPR
42%monkey oral F

A thousand-fold weaker, and finally absorbed by mouth.

ExchangeCompound 18 µM
SPRCompound 10<1 pM
NHP FCompound 2342%
over wild type 55×compound 5a, SPR
KRAS 2 / 3
Sheet16/25PaperKRAS G12DCpd23PDB9E5D + 9E5F
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/kraspapermdeller.com/krasdoi

The structure answered a metabolism question, not a binding one.

J. Med. Chem. 2026, 69, 25112-25127Discovery of INCB161734: A Selective Inhibitor of the KRAS G12D MutantSokolsky et al.10.1021/acs.jmedchem.6c02625

The discovery of the clinical KRAS G12D inhibitor INCB161734 is described. Starting from a previously disclosed orally bioavailable tool compound 1, a concerted effort to improve the predicted human dose via modulation of potency and pharmacokinetics was undertaken. Careful consideration of the metabolic profile was key to designing low clearance compounds. In addition, the molecular size and fraction sp3 character were carefully controlled to improve permeability and crystalline solubility. From this effort, INCB161734 was identified as a balanced compound with excellent oral exposure, potent antiproliferative activity in xenograft tumor models, and excellent selectivity against wild-type K/H/NRAS. INCB161734 is currently undergoing clinical trials in patients with KRAS G12D mutations.

The shunt, closed by shapemorpholinoneshuntscaged amine cleavedpyrrolidineno cleavage seenpredicted human clearanceabove 1.1 L/h/kgbelow 0.5
Static render of this view

Permeability against clearance, every compound

364 nMwhole-blood pERK
485×over wild type

One residue carries the whole RAS family.

startINCB159020, the tool compoundh-Cl 1 L/h/kg
the shuntPyrrolidine for morpholinoneno cleavage seen
one atomMethyl at the pyrrolidine 5-positionh-Cl >1.1 → <0.5 L/h/kg
candidateINCB161734monkey F 76%
KRAS 3 / 3
Sheet17/25PaperKRAS G12DCpd35PDB7RPZ
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismpapermdeller.com/krasbdoi

Inhibit JAK2 and you take the blood counts with you.

J. Med. Chem. 2026Discovery of the Orally Bioavailable Isoform Selective Janus Kinase 1 (JAK1) Compound Povorcitinib (INCB054707) for the Treatment of Inflammatory and Autoimmune DiseasesZhuo et al.10.1021/acs.jmedchem.5c03753

Aberrant JAK2 signaling is a key driver in cancers such as myeloproliferative neoplasms (MPNs), whereas JAK1 acts as a central mediator of autoimmune and inflammatory diseases. Since JAK2 supports blood cell production, its inhibition can lead to anemia and other cytopenias, making selective JAK1 inhibition a potentially safer therapeutic strategy opposed to dual JAK1/JAK2 inhibition. We initially identified novel scaffold 9, which selectively inhibited JAK1 over JAK2. Iterative breakthroughs led to the discovery of povorcitinib (INCB054707, 22), which showed improved pharmacokinetics due to intramolecular hydrogen bonding of the amide moiety and a reduced ring size in the hinge binding motif. Protein crystallography suggested that JAK1 selectivity was conferred by the amide's alkyl group accessing a small lipophilic groove formed by the P-loop conformation of JAK1 that was distinct from JAK2. In murine arthritis models, povorcitinib demonstrated dose-dependent efficacy and reduced inflammatory signaling, supporting its therapeutic potential in immunological disorders.

OUTINcytokine receptorJAK1JAK2hitspareSTATred cells andplateletsinflammation runs through JAK1JAK1 vs JAK2
Static render of this view

Povorcitinib holds the JAK1 hinge with two hydrogen bonds.

Hinge Glu957 · Leu959pyrazole N-H and N, both measured2.8 Å · 2.9 Å
Gatekeeper Met956same pose in JAK2 (10PJ)
Glycine loopGly884 and Gly887 backbone3.5 Å · 3.4 Å
Water to His885one bridging water2.9 Å
Water bridgeone ordered water, drawn dotted
Hydrophobic contactsfaint and dotted, behind the polar picture
Selectivity is not in this picturetake-home
Rat Fortho-F+2×
Cyno AUCCF3 amide+37×
JAK1pyrazole−1,256×
Dog Fmethyl+100×
JAK1 1 / 2
Sheet18/25PaperJAK1Cpd22PDB10PI
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/jak1papermdeller.com/jak1doi

Same ligand, same hinge, in both isoforms.

Static render of this view

JAK1 vs JAK2 IC50, every compound (nM)

Oral bioavailability in three species

11 nMJAK1 IC50
46×over JAK2

Selectivity from a P-loop 2 Å higher; forty-four-fold on the enzyme bought full absorption.

JAK1Compound 91.5 nM
JAK1Compound 190.25 nM
Rat FCompound 22>100%
JAK1 2 / 2
Sheet19/25PaperJAK1Cpd22PDB10PI + 10PJ
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismblogmdeller.com/jak1papermdeller.com/jak1doi

Five campaigns, each one decided by reading the structure first.

FGFR3≠FGFR142×
compound 29number not disclosednew chemotype, following pemigatinib · similarity 0.13no registered trial
CDK2≠CDK1220×
compound 17programme: INCB123667, phase 3new chemotype, against palbociclib · similarity 0.16NCT07214779recruiting
KRAS G12D≠KRAS WT55×
INCB159020 (23)programme: INCB161734, phase 3new chemotype, against sotorasib · similarity 0.18NCT07522073recruiting
JAK1≠JAK246×
INCB054707 · povorcitinibphase 3new chemotype, following ruxolitinib · similarity 0.18NCT05620836completed
mutCALR≠CALR WTno WT epitope
Fab1 FvWO 2023/107994an antibody, not a small molecule · solved in 10 dayspublished 15 June 2023
Take homestructure firstSolve the complex, read every contact, then design against the difference between the target and the thing you must not hit.
Summary
Sheet20/25Paperall fiveCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/publicationssitemdeller.com/structures

One loop, and how far it carried each time.

DesignMakeTestCrystalStructureSBDDOUTFGFR2 kinase domain · 8E1XWITH COMPOUND 30 · THE CHEMOTYPE THAT HOLDS THE MUTANTSMARGININDPHASE 1PHASE 3FGFR3 · compound 2942× over FGFR1no registered candidateCDK2 · compound 17220× over CDK1INCB123667†PHASE 3, RECRUITINGKRAS G12D · compound 2355× over wild typeINCB161734†PHASE 3, RECRUITINGJAK1 · compound 2246× over JAK2povorcitinibPHASE 3, COMPLETEDmutCALR · the Fvno wild-type epitopeWO 2023/107994 filed, 10 days
Static render of this view
The loop
Sheet21/25Paperall fiveCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/publicationssitemdeller.com/structures

Three companies. One move: designing next-generation therapeutics.

Elora Therapeutics

Co-founder and CSO

Engineered enzyme medicines for microplastic-associated disease.

AN ENZYME AND A POLYMERPETEDCBA12345IIIPETasenickEDCBAnew end12345new endABCDEIII54321IIIMHET, then TPA and ethylene glycol

DeepCovalent

Scientific technical adviser

De novo small molecules designed in a single pass.

A MODEL AND THE PHYSICSOS3Done passphysicscofold, MD, free energysynthesis-ready

Arcadus Therapeutics

Building the ARC Engine

Programmable induced proximity: designing the ternary complex.

A LIGASE AND A PROTEINCRBNE3 ligase9FJX, 2.00 Åtargetno pocketdesigned glueubiquitin
Building now
Sheet22/25Paper·Cpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanism

Eighteen live apps, and sixteen of them answer a stage of this work.

PANTSpick the targetFAFFAbout·TopPDBLXcrystalliseCODSWALLOP·BINMANALPHABETTIthe archiveBoltzMaker·GOBSMACKEDAlphaFraudpredictChemSagechemistryFlexAppeal·ButtFolddynamicsChatPDB·chatMCDGATECRASHER·PODIUMexplain it
TRAINChemSage
ChemSage32B · 4-bit · rsLoRA r64

Calculate the TPSA for CN(Cc1ccc(F)cc1)S(=O)(=O)c1ccc(C)cc1.

from rdkit import Chemfrom rdkit.Chem import rdMolDescriptorsmol = Chem.MolFromSmiles('CN(Cc1ccc(F)cc1)S(=O)(=O)c1ccc(C)cc1')print(round(rdMolDescriptors.CalcTPSA(mol), 1))

TPSA = 37.4 Ų. This compound meets the Veber oral absorption threshold (TPSA ≤ 140 Ų).

round 5 test set, unedited · 9.65 s

base
Qwen2.5-32B, 4-bit
adapter
rsLoRA r64, 268M trainable
data
20,000 examples, 78 classes
eval
95% overall, against 80% at v4
SIMULATEFlexAppeal
system
GLP-1R + orforglipron, 6XOX
run
amber14, 0.200 ns, 56 ns/day
RMSD
1.80 Å, last 10% of frames
site
grade A, 96.7 / 100
LISTENButtFold
chain
ubiquitin, 76 residues
poses
150, 50,000 steps apart
compaction
Rg 21.3 to 11.5 Å
fidelity
Q 0.973
Vibe coding
Sheet23/25Papergithub.com/bellcheddar
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanism

Challenges and themes in targeted precision oncology.

Targeting RAS proteins beyond G12C and G12D

KRAS, PDB 6GOD, molecular surface with residues 12, 13 and 61 in red and the bound GppNHp as sticks
NRAS, PDB 5UHV, molecular surface with residues 12, 13 and 61 in red and the bound GppNHp as sticks
HRAS, PDB 5P21, molecular surface with residues 12, 13 and 61 in red and the bound GppNHp as sticks
KRAS 19.3%NRAS 5.3%HRAS 2.3%6GOD · 5UHV · 5P21, surfaces, sized by frequencyof every RAS-mutant cancerKRAS   71.7%NRAS   19.7%HRAS   8.6%

Targeting the dark kinome

dark kinases in blue119unexplored62 dark131underexplored58 dark137explored29 dark131clinical6 dark

Targeting pseudokinases

outinFFSS2211held offWTV617FFFSS2211released, and firingFERMSH2JH2pseudokinaseJH1kinaseevery approved drug
Still open
Sheet24/25Paperthree reviewsCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanismsitemdeller.com/publicationssitemdeller.com/structures

Thank you for your time.

Hit this. Spare that.

numerous times, over twenty years

FGFR342×FGFR1
CDK2220×CDK1
KRAS G12D55×KRAS WT
JAK146×JAK2
mutCALRno WTCALR WT
400+ structures5 campaigns3 in phase 3

Leeds · Oxford · Yale · Pfizer · Scripps · Stanford · Incyte

Sheet25/25Paperall fiveCpd·PDB·
PODIUM: Presentation Of Drug-discovery Impact, Understanding & Mechanism