Absci Corporation · NASDAQ:ABSI
Absci — Investor Presentation (August 2025)
74 slides from Absci's investor presentation, Aug 12, 2025
Slide 1: Corporate Presentation Summer 2025
from absci import de_novo_model model = de_novo_model.load_latest() antigen = model.load_pdb("7olz.pdb", chain="A") antibodies = model.predict(antigen, N=300000) from absci_library import codon_optimizer library = codon_optimizer.reverse_translate(library) library.to_csv("covid-antibody-designs.csv"
Slide 2: Disclaimers
2C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Disclaimers Forward-Looking Statements Certain statements in this presentation that are not historical facts are considered forward-looking within the meaning of Section 27A of the Securities Act of 19
Slide 3: Data-first generative AI for Drug Creation
3C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Data-first generative AI for Drug Creation™ P I P E L I N E W I T H N E A R - T E R M C A T A L Y S T N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B C I C O R P O R A T I O N . A L L R
Slide 4: Amassing high-quality data at scale for AI model training and validation
4C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Since 2020 Absci has been amassing high-quality data at scale for AI model training and validation DATA TO TRAIN Proprietary High throughput screening assays generate high- quality data for generative
Slide 5: Leadership in AI de novo design of antibody-based therapeutics
5C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Leadership in AI de novo design of antibody-based therapeutics EVQLSEVGA . . . de novo antibody design model creates epitope-specific binders given a target structure Designed in framework of choice or
Slide 6: We use AI to create novel & differentiated therapeutics
6C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ENABLING FEATURES: MULTI-VALENCY, pH-DEPENDENT BINDING ABILITY TO ADDRESS DIFFICULT TARGET CLASSES, E.G. GPCRS EPITOPE-SPECIFIC DESIGN + EPITOPE INTERFACE OPTIMIZATION BROAD IP: 100S TO 10,000S OF FUNC
Slide 7: Leveraging AI throughout the end-to-end drug discovery process
7C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Leveraging AI throughout the end-to-end drug discovery process I N T E G R A T E D D R U G C R E A T I O N P L A T F O R M TARGET SELECTION AI reverse immunology target discovery OR Additional Target S
Slide 8: de novo model progress 2022-2025
8C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . de novo Model v1 Absci was the first to design and validate novel antibodies using zero-shot generative AI in BioRxiv preprint de novo Model v2 Demonstrated de novo design model’s broad applicability t
Slide 9: Platform Case Studies
9C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Platform Case Studies AI LEAD OPTIMIZATIONDE NOVO ANTIBODY DESIGN Goal: create universally neutralizing HIV antibody by binding conserved epitope within “caldera” region of HIV gp120 Absci’s de novo de
Slide 10: Multilingual team with expertise in AI and drug creation
10C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ”Multilingual” team with expertise in AI and drug creation Sean McClain Founder, CEO & Director Andreas Busch, PHD Chief Innovation Officer Zach Jonasson, PHD Chief Financial Officer & Chief Business
Slide 11: Talent and Infrastructure for Better Biologics Faster
11C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ~150 77,000+ Square Feet >$600M Unlimiters with deep experience in AI, drug discovery, immunology, and synthetic biology State-of-the-art drug creation and wet lab space in Vancouver WA, Absci AI Rese
Slide 12: Advancing and expanding our pipeline of novel & differentiated assets
12C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . INFL. Bowel Disease / TL1A Androgenetic Alopecia / PRLR Immuno-oncology / Undisclosed Additional Early Discovery Programs Target Lead ID Candidate ID IND-Enabling ABS-101 ABS-201 ABS-301 Phase 1 Oncol
Slide 13: Track Record of Industry-Leading Partnerships
13C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . AI Drug Creation™ Data & compute 25+ PARTNERED PROGRAMS TO DATE SCALING COMPUTE 4 NAMED INTERNAL PROGRAMS ADDITIONAL PROGRAMS IN EARLY DEVELOPMENT IMPROVING MODELS INCREASING EFFICIENCIES Trademarks,
Slide 14: Near-term value inflection points
14N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Leading AI platform driving numerous near-term value inflection points A B S - 1 0 1 Phase 1 Study initiated – May 2025 Phase 1 Interim Data Readout – anticipated 2H 20
Slide 15: Absci's progress in Drug Creation
15N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . A B S - 1 0 1 Phase 1 Program initiated in May 2025 • New preclinical data supporting superior immunogenicity profile • Phase 1 Interim data readout expected 2H 2025 A
Slide 16: Validated mechanism of action in large underserved market
16N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Validated mechanism of action in large underserved market A B S - 1 0 1 T L 1 A P O T E N T I A L R E L E V A N C E I N W I D E R A N G E O F A U T O I M M U N E I N D
Slide 17: Potential best-in-class TL1A mAb designed using generative AI
17N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Potential best-in-class TL1A mAb designed using generative AI ABS-101 designed to achieve superior therapeutic properties and differentiation over first generation clin
Slide 18: AI platform generates high affinity variants
18N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Successful application of AI platform to generate high affinity variants A B S - 1 0 1 T L 1 A H I G H A F F I N I T Y m A b s W I T H P R E S E R V E D C R O S S - R E
Slide 19: High affinity mAbs with binding to TL1A monomer and trimer
19N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . 1101001000 1 10 100 1000 Monom er Bind ing (pM) Tr im er B in di ng (p M ) No M onomer Bind ing H I G H A F F I N I T Y m A b s W I T H B I N D I N G T O B O T H T H E
Slide 20: Immunogenicity assessment of mAb:TL1A complexes
20N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Internalization of mAb:TL1A complexes potentially contributes to immune activation and formation of ADA A B S - 1 0 1 T L 1 A | I M M U N O G E N I C I T Y A S S E S S
Slide 21: NHP pharmacokinetics & CMC data
21N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . NHP Pharmacokinetics & CMC data confirm compelling ABS-101 competitive profile A B S - 1 0 1 T L 1 A | N H P C O M P A R A T I V E P K 2-3x extended half-life in NHPs o
Slide 22: NHP data confirm sustained target engagement
22N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-101 NHP data confirm sustained prolonged targe engagement versus clinical competitors A B S - 1 0 1 T L 1 A | N H P C O M P A R A T I V E P D ! "! #! A! %! &! '! !(
Slide 23: AI-designed for potentially optimal therapeutic profile
23N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . AI-designed for potentially optimal therapeutic profile A B S - 1 0 1 T L 1 A A T T R I B U T E ABS - 1 0 1 M K - 7 2 4 0 ( M E R C K , P R O M E T H E U S ) R V T - 3
Slide 24: Phase 1 trial initiated May 2025
24N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Phase 1 Clinical Trial Initiated in May 2025 with interim readout expected 2H2025 A B S - 1 0 1 T L 1 A AI-designed Development Candidate ü High affinity ü High potency
Slide 25: Absci's progress in Drug Creation
25N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . A B S - 1 0 1 Continued progress of TL1A asset with FiH in 1H 2025 • New preclinical data supporting superior immunogenicity profile • Phase 1 Interim data readout in 2
Slide 26: ABS-201 potential to unlock new category in hair re-growth
26N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-201 has the potential to unlock a wholly new category of therapy in hair “re-growth” Significant unmet clinical need for androgenetic alopecia Large market: approxi
Slide 27: Underserved patient population
27N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Underserved patient population looking for therapeutic innovation FEMALE ANDROGENETIC ALOPECIA ~30M women in the U.S. Only 1 FDA approved therapy for women ~50M men in
Slide 28: Patients need better treatment options for hair re-growth
28N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Patients and clinicians need better treatment options for “hair re-growth” • Hair re-growth, not just slowing of hair loss • Safe • Minimal side effects • Durable effec
Slide 29: Potential best-in-class PRLR antibody
29N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Potential best-in-class PRLR antibody for treating androgenic alopecia A B S - 2 0 1 P R L R AI-designed PRLR program for durable hair regrowth therapy addressing patie
Slide 30: PRLR inhibition as innovative alternative
30N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . PRLR inhibition as a safe innovative alternative to current treatment options A B S - 2 0 1 P R L R | P R O P O S E D D I R E C T M O D E O F A C T I O N P r o p o s e
Slide 31: Prolactin impacts on organ-cultured human hair follicles
31N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Prolactin impacts on organ-cultured human hair follicles 1doi: 10.2353/ajpath.2006.050468 P r o l a c t i n - d r i v e s h a i r f o l l i c l e r e g r e s s i o n i
Slide 32: PRLR inhibition supported by human genetics
32N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . PRLR inhibition anticipated to be safe & well tolerated as supported by human genetics A B S - 2 0 1 P R L R I. II. III. NV = Nonvariant ND = Not determined I. II. III.
Slide 33: Anti-PRLR mAb promotes long-term hair growth in NHP
33N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . TREATMENT WITH AN ANTI-PRLR MAB PROMOTES AND SUSTAINS LONG-TERM HAIR GROWTH IN NHP Hair density & thickness improved with short treatment duration in primate model of a
Slide 34: ABS-201 candidates are high affinity and potent binders
34N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-201 candidates are high affinity and potent binders A B S - 2 0 1 P R L R | I N V I T R O P R O F I L E PathHunter - in vitro PRLR reporter functional assay 10 - 3
Slide 35: ABS-201 superior efficacy vs minoxidil
35N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-201 shows superior efficacy vs 5% topical minoxidil in 21d hair regrowth model Administration: mAbs i.p. biweekly; Minoxidil topical daily Untreated (n=11) Isotype
Slide 36: NHP pharmacokinetic data for ABS-201
36N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . 56 day interim NHP pharmacokinetic data confirms ABS-201 is well positioned for AGA market A B S - 2 0 1 P R L R >3x extended half-life in NHPs compared to HMI-115 High
Slide 37: Modelling shows superiority of ABS-201 vs HMI-115
37C O N F I D E N T I A L A N D P R O P R I E T A R Y C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Modelling shows superiority of ABS-201 vs HMI-115 on PK & Receptor Occupancy A B S - 2 0 1 , P R L R | I N S I L I C O M O D E L I N G >3x extended
Slide 38: Superior profile of ABS-201
38C O N F I D E N T I A L A N D P R O P R I E T A R Y C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . D E S I R E D A T T R I B U T E HMI-115 ABS-201 A F F I N I T Y + ++ I N V I T R O P O T E N C Y ++ ++ H I G H S O L U B I L I T Y - ++ S T A B I L
Slide 39: Straightforward path for ABS-201 clinical development
39N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Straightforward path for ABS-201 clinical development C L I N I C A L T R I A L S F O R H A I R T R E A T M E N T S A R E E X P E C T E D T O B E S T R A I G H T F O R
Slide 40: Scientific Advisory Board of Hair Experts
40N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Leading Scientific Advisory Board of Hair Experts DR. ANTHONY ROSSI Memorial Sloan Kettering Cancer Center DR. CHESAHNA KINDRED Kindred Hair & Skin Center DR. MATT L. L
Slide 41: Absci's progress in Drug Creation
41N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . A B S - 1 0 1 Continued progress of TL1A asset with FiH in 1H 2025 • New preclinical data supporting superior immunogenicity profile • Phase 1 Interim data readout in 2
Slide 42: Tertiary lymphoid structures and antibodies
42N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . The presence of TLS is associated with longer progression-free survival and better response to immune checkpoint inhibitors2,3. Rapidly growing evidence illustrates cor
Slide 43: Reverse Immunology platform identifies antigens
43N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . LIGHT CHAINS HEAVY CHAINS Samples collected from patients with high expression of TLS markers Antibody expression Immunoglobulin reads from RNAseq data Target identific
Slide 44: ABS-301 blocks an immunosuppressive cytokine
44N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . A B S - 3 0 1 r e s c u e s p r o - i n f l a m m a t o r y s i g n a l i n g t h r o u g h i n h i b i t i o n o f i m m u n o s u p p r e s s i v e c y t o k i n e A
Slide 45: In vivo target validation: anti-tumor response
45N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . In vivo Target Validation: Pathway Activation Drives Potent Anti-Tumor Response A B S - 3 0 1 Key Findings: Activation of the ABS-301–targeted pro-inflammatory pathway
Slide 46: Target expression across squamous cell carcinomas
46N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Expression of ABS-301’s target suggests broad potential in squamous cell carcinomas A B S - 3 0 1 Distribution of ABS-301 target expression across squamous cell carcino
Slide 47: ABS-301 expression in lung squamous cell carcinoma
47N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-301 | Expression in Lung Squamous Cell Carcinoma (LUSC): no change with treatment and strong negative correlation with CD8+ T cell infiltration In LUSC, univariate
Slide 48: Broad potential in immuno-oncology
48N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-301 | Broad potential in immuno-oncology Based on literature and potential competitive molecules, the following indications could be of interest: *dependent on stag
Slide 49: Absci's progress in Drug Creation
49N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . A B S - 1 0 1 Continued progress of TL1A asset with FiH in 1H 2025 • New preclinical data supporting superior immunogenicity profile • Phase 1 Interim data readout in 2
Slide 50: De novo AI model on HER2 discovers distinct antibodies
50N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . • Hits with edit distance of up to 12 amino acids in HCDR3 region (13 aa, search space of 2013) were screened • Selected 50 hits with <10 nM affinity were expressed as
Slide 51: AI-designed antibodies enhance ADCC activity vs trastuzumab
51N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Luciferase signal driven by NFAT transcription factor positively correlates to ADCC activation against JIMT-1 ABS-501, HER2 | AI-designed antibodies demonstrate measura
Slide 52: AI-designed antibodies suppress HER2+ breast tumors
52N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-501, HER2 | AI-designed antibodies suppress growth of trastuzumab-sensitive & resistant HER2+ breast tumors Partial Critical Not critical Trastuzumab WT Variant 1 V
Slide 53: AI-designed antibodies address unmet medical need
53N O N - C O N F I D E N T I A L C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . ABS-501, HER2 | AI-designed antibodies create opportunities to address unmet medical need Modality switch or combination opportunities under consideration to address un
Slide 54: Leading AI models to create novel & differentiated therapeutics
54C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Leading AI models to create novel & differentiated therapeutics “Smart” biologics Enhanced Potency & MOA Engineer selectivity, minimizing off target toxicity Agonism vs. Antagonism Bind Specific extra
Slide 55: Leadership in AI de novo design of antibody-based therapeutics
55C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Leadership in AI de novo design of antibody-based therapeutics EVQLSEVGA . . . de novo antibody design model creates epitope-specific binders given a target structure Designed in framework of choice o
Slide 56: AI platforms and lab-in-the-loop cycles
56C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . AI Platform designs antibodies Wet Lab confirms AI- designed antibodies maintain drug-like properties Wet Lab data improve models D E N O V O A N T I B O D Y D E S I G N Our AI platforms are enabled b
Slide 57: AbsciDesign AI models for de novo antibody design
57C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . AbsciGen: antibody<>antigen complex structure and sequence design AbsciBind: antibody design scoring and filtering Antigen AbsciGen AbsciBind AbsciBind High Rank RMSD = 2.3 Å Confidence = 0.95 AbsciBi
Slide 58: End-to-end design-validation workflow
58C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . STEP 1. Define design parameters STEP 2. Fine-tune and deploy AbsciGen and AbsciBind to generate hundreds of thousands of variants and filter to a subset that are likely binders STEP 3. Wet lab screen
Slide 59: Case study: de novo design of antibody binding HIV-1 Caldera
59C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . C A S E S T U D Y de novo design of an antibody that binds the Caldera region of HIV-1 trimer
Slide 60: No natural or synthetic antibody universally neutralizes HIV
60C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . No natural or synthetic antibody for HIV exists today because immune system cannot derive an antibody that is universally neutralizing against HIV Design challenge: create universally neutralizing HIV
Slide 61: HIV-Caldera: Determine inputs and design
61C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . HIV-Caldera: Determine inputs and design D E N O V O D E S I G N HIV Env Trimer Challenge : • Highly glycosylated • Extremely high sequence diversity among isolates • High mutation rate at common neut
Slide 62: Best structures selected from 10,000+ generated
62C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . 4 best structures selected from 10,000+ structures generated by de novo model D E N O V O D E S I G N 17b Structure S1 HCDR3 Structure S3 HCDR3 Structure S2 HCDR3 Structure S4 HCDR3 HeavyLight
Slide 63: Molecular dynamics simulation of designed antibodies
63C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Applied molecular dynamics simulation to de novo designed antibodies D E N O V O D E S I G N
Slide 64: Enriched de novo library binds open Env trimer
64C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . H A ta g Antigen Clade A Env trimer Closed Open Enriched de novo library binds open, not closed, Env trimer conformation in YSD D E N O V O D E S I G N Closed Open Clade B Env trimer
Slide 65: SPR data consistent with caldera binding
65C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . SPR data demonstrate binding characteristics consistent with binding of caldera D E N O V O D E S I G N
Slide 66: SPR shows no binding to GP120 monomer
66C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . HIV-Caldera: SPR demonstrates no binding of de novo designs to GP120 monomer D E N O V O D E S I G N Hypothesis: If the designed mAbs are binding to the caldera region we should not observe binding to
Slide 67: HIV-Caldera: AI de novo design for challenging target
67C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . HIV-Caldera: demonstrating AI de novo design for challenging target H I V D E N O V O D E S I G N SUMMARY de novo design model created a novel and diverse antibody which binds multiple clades of HIV i
Slide 68: Case study: AI optimization for pH sensitivity
68C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . CASE STUDY A I O p t i m i z a t i o n f o r p H s e n s i t i v i t y
Slide 69: AI lead optimization for pH sensitivity
69C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . C A S E S T U D Y – A I L E A D O P T I M I Z A T I O N f o r p H S E N S I T I V I T Y AI lead optimization platform for ‘smart biologics’ T H E C H A L L E N G E : The diversity of antibodies is vas
Slide 70: Tumor specificity improves efficacy and reduces toxicities
70C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . TUMOR SPECIFICITY IMPROVES EFFICACY AND REDUCES ”ON-TARGET OFF-TUMOR” TOXICITIES pH sensitivity may reduce toxicity and/or improve efficacy of therapeutic mAbs C A S E S T U D Y - A I L E A D O P T I
Slide 71: Models identify pH sensitive Fab variants
71C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Models identify pH sensitive Fab variants from the same lead for either indication C A S E S T U D Y - A I L E A D O P T I M I Z A T I O N f o r p H S E N S I T I V I T Y 1. Library for model training
Slide 72: Hits reformatted as mAbs show desired binding profiles
72C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . Hits reformatted as mAbs show desired binding profiles C A S E S T U D Y - A I L E A D O P T I M I Z A T I O N f o r p H S E N S I T I V I T Y AI optimized leads achieves variants with pH sensitive bi
Slide 73: Summarized platform case studies
73C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D . D E N O V O D E S I G N de novo design model created molecule binds multiple clades of HIV suggesting successful targeting of the caldera epitope Represents second disclosed target success for our de
Slide 74: Better biologics for patients, faster
Better biologics for patients, faster 74C O P Y R I G H T © 2 0 2 5 A B S C I C O R P O R A T I O N . A L L R I G H T S R E S E R V E D .