FinalDose is building the first programmable drug platform - a single smart drug molecule that finds diseased cells by their DNA and destroys them. They're starting with all cancers. Congrats on the launch, Jeff Liu, Steven Lin, and @liyaohuang2!

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Now can we have the real Biotech version of this please? Why are YC Bio pitches becoming so dumbed down that it’s impossible to believe them as a biologist!

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Good luck presenting yourself to big pharma like this

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Congrats on funding and launch. Now, instead of all cancers, just do KRAS G12X. That will show the real moat. And then account for subclonal evolution, epigenetic changes, and tumor heterogeneity.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 This is stupid. We give all the money to fake grifters. YC and VCs are cooked. There are plenty of cancer bio start ups working hard to produce companies that at least make sense. Either they are explaining it wrong or they just skipped a ton literature on cancer bio.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Constructive criticism: none of this makes sense.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Wow respect to YC not another B2B SaaS wrapper. Maybe there is hope after all.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Scam SV biotech

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Ignore the comments about it not being "technical enough". Had it been that ppl will complain it felt like a biotech webinar. Can't satisfy everyone. Launch videos should be broad enough to be accessible by laymen too. Congrats on the launch and going after cancer..

feasibility for anyone interested via The core concept is scientifically grounded in real research, but the specific claims made in this pitch contain several significant overstatements that any scientist or investor should scrutinize carefully. What's Real and Supported by Evidence The idea of programmable, DNA-sensing molecules that trigger cell death is not science fiction. DNA nanotechnology has produced nanostructures with "unprecedented programmability" and "intrinsic biocompatibility" that can interact with cancer cells and serve as therapeutic agents [1]. Nucleic acid-based functional nanomaterials (NAFN) with "inherent biological functions and sequence programmability as well as biocompatibility and biodegradability" have been demonstrated in chemo-, immuno-, and gene-therapy applications [2]. RNAi-based platforms already achieve programmable, sequence-specific silencing, and researchers are actively developing "ligand-guided targeting" and "stimuli-responsive release" systems [3]. CRISPR-based approaches have already demonstrated that delivering guide RNAs with Cas9 can induce cell-type-specific death in cultured cells, in vivo xenografts, and patient-derived cancer organoids [7]. G-quadruplex aptamers, which are DNA-based molecules, have shown programmable cytotoxicity in cancer cells at low-nanomolar potency [8]. Synthetic gene circuits and engineered immune cells like CAR-T constructs can "selectively recognize tumor-associated antigens and overcome immunosuppressive barriers" [5]. So the underlying biology is real. Where the Pitch Overstates Reality "Cure all cancers" with one platform is the central problem. Cancer is not one disease. It is hundreds of diseases with enormous genetic heterogeneity, not just between cancer types but between patients with the same cancer type, and even between cells within a single tumor. A sequence-targeted kill switch that works against a MYC amplification in one patient's lung cancer may be completely irrelevant to another patient's MYC-driven tumor if the driving mutation is in a different region, or if the tumor has already evolved resistance by deleting or mutating the target sequence. Tumor evolution and clonal escape are well-documented phenomena that have defeated many "precision" approaches. "Neighboring healthy cells are untouched" is the hardest claim to defend. This is the central unsolved problem in oncology. The target DNA sequences driving cancer (MYC overexpression, TP53 loss-of-function mutations, APC truncations) are often present in normal cells at low levels, in germline tissue, or in non-cancerous cells that share the same mutation. Off-target effects of DNA-targeting tools are a recognized and serious safety concern, and therapeutic gene editing safety is "intricately linked to the specificity of the editing tools used to cut at precise genomic targets," with off-target editing remaining a major unresolved challenge [12]. The AID/APOBEC family of DNA-modifying enzymes, for example, can cause genomic instability in cancer but also in normal cells [4], illustrating how difficult it is to confine DNA-level interventions to diseased tissue only. "R&D cycle and cost collapses to near 0" is marketing language, not science. Even if the molecular design is modular, every new guide sequence targeting a new cancer mutation requires independent preclinical validation, toxicology studies, delivery optimization, and clinical trials. The FDA's regulatory pathway for gene therapies and novel biologics does not compress because the underlying platform is elegant. The FDA's Accelerated Approval Program exists precisely because even promising platforms require demonstrated clinical benefit [17]. Delivery is the unaddressed elephant in the room. Getting a large, charged molecule like a DNA-sensing protein or nucleic acid construct into the nucleus of a cancer cell, across every tumor type, in every tissue, without triggering immune responses or being degraded, is one of the hardest problems in the entire field. RNAi therapeutics have faced exactly this barrier for decades, and "clinical translation hinges on overcoming instability, immunogenicity, and inefficient endosomal escape" [3]. The pitch does not mention delivery at all. "85% of disease drivers have no druggable protein" is a real and widely cited observation in the field, and the frustration with protein-centric drug discovery is legitimate. But the implication that DNA-level targeting cleanly solves this is a leap. Many "undruggable" targets are undruggable not just because they lack a binding pocket, but because of redundancy, tissue distribution, and the difficulty of achieving therapeutic windows. Bottom Line The scientific concepts FinalDose describes, including programmable DNA-sensing molecules, conformational kill switches, and sequence-specific cell death, are real areas of active research with published proof-of-concept data [1][2][7][8]. What is not supported is the claim that this constitutes a single platform capable of curing all cancers with negligible R&D cost and perfect selectivity. No such platform exists in the peer-reviewed literature, and the unsolved problems of tumor heterogeneity, off-target effects, delivery, and immune response are not acknowledged anywhere in the pitch. This reads as a scientifically literate but heavily promotional early-stage startup pitch, not a description of a validated therapeutic platform. The concept is scientifically plausible at the level of individual proof-of-concept experiments, but the "cure all cancers" framing is not supported by any existing evidence and glosses over the hardest unsolved problems in the field.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 As a molecular biologist: How do I short this?

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 What’s the delivery strategy and PK? Does it target endocytosis by all cells then migrate to the nucleus and identify if this cell is a target or does it know extracellularly somehow? Is this some kind of targeted masked crispr or something?

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Is this the level needed to bag millions of dollars for a “startup”? Seriously?

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Whether it works or not, this should inspire more people. Cancers are actually nightmare, hopefully it becomes a past.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 ignore the snooty biotech replies. shoot for the stars chaps - you might just land on the moon

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Good job! LFG

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 this is awesome

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Congrats @sklin_lite. Great launch. This will be a win for humanity

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Nice

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 congrats on the launch, very amazing work guys

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 congrats guys, incredible stuff

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 congrats y'all! meaningful progress for all of humanity, excited to see where this goes

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Congrats on launch! Had great conversations with them. Looking forward to what you guys will do in the future!

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 These guys are even imitating the haircuts of Feng Zhang and David Liu

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Congratulations!

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Sick guys! Well done.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 This is cool! Congrats guys!

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 this is so tuff 🐐

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 congrats guys! coolest thing i've ever seen

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Two new publications showing something very like this (2nd article in next msg). V glad to see that directly targeting mutations in cancer is getting traction. Ultimately it's the only way IMO.

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Congrats! So excited for this

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 the hard part isn't writing code or even iterating fast, it's finding diseased cells by their dna and not killing healthy ones in the process, which is a monumental problem to solve in the first place

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 SIVARR Is buidling an AI that serves as your personal assistant and knows you and your company better than you. Are you a founder, content creator, builder or you make use of 2-35+ different apps for productivity. Sivarr is here for you Follow for more update on SIVARR. 🚀

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 Congrats!

@Jeffliu6068Liu @sklin_lite @LiYaoHuang2 I have a disease that affects 1 in 500 people - more common than multiple sclerosis, lupus, or Marfan syndrome yet no one knows what causes it, could you apply this tech @Jeffliu6068Liu