AI cannot find drugs curing cancers.

Home page
AI is hyped, useless
AI cannot predict drugs or superbugs

AI, Alphafold's bad success rate of predicting proteins bound to cancer cells.  ← No cure for diseases.

(Fig.1)  Overhyped AI's prediction rate of proteins is very bad = only 2%

AI cannot design proteins binding to cancer cells.

AI cannot design binder proteins against target cancer antigens (= NY-ESO-1 ) for CAR-T-cell therapy.

The 7th, 2nd-last paragraphs of this hyped news (7/24/2025) say

"a well-known cancer target, NY-ESO-1, which is found in a wide range of cancers. The team succeeded in designing a minibinder that bound tightly to the NY-ESO-1 pMHC molecules. When the designed protein was inserted into T cells,.. which effectively guided the T cells, to kill cancer cells in laboratory experiments (= Not inside bodies, so still useless )."

"expects that it will (= still useless ) take up to five years before the new method is ready for initial clinical trials in humans."

T cell treatment of cancers, AI are deadend.

This research tried to aim at already-deadend CAR-T cell therapy against cancers with already-known cancer antigen in vain.

↑ So this research tried to make AI, Alphafold design minibinder proteins (= with very low success rate of less than 2%, this-p.8-1st-paragraph, which is useless, too bad to predict side effects ) binding to cancer antigen to stimulate T cell to kill cancers in vitro (= Not inside animals or bodies ).

↑ This CAR-T cell therapy for solid tumors did Not succeed despite many attempts, because T-cells are often blocked from reaching cancer cells inside bodies ( this-3.car T-cell theory in solid tumors,  this-p.1-left-last-paragraph ).

This-abstract says  -- Deadend T-cell therapy
"Despite early signs of clinical efficacy, CAR-T cells have repeatedly failed to achieve curative responses in solid cancers."

AI, Alphafold are useless, cannot predict proteins.

AI, Alphafold rates of predicting proteins binding to target cancer cell's antigens were very bad = only 2 ~ 10%, which cannot predict side effects.

↑ This research paper ↓

p.2-last-paragraph says  -- AI low success rate
"Among various de novo protein design approaches, RFdiffusion (= AI trained on experimental protein structure in PDB, this-p.68-4.1, Not quantum mechanics ) is considered one of the most promising tools due to an impressive average of ~10% success rate of designed binders (= too bad and low AI success rate )"

p.3 says  -- AI's design of binder to cancers
"In our work, we focused on designing binders that specifically target the pMHC formed by MHC-I, presenting the cancer-associated NY-ESO-1 peptide"
"Following backbone generation through RFdiffusion denoising trajectories, we designed sequences using ProteinMPNN. The resulting designs were filtered based on AF2 (= Alphafold 2 ) initial guess"

p.5-1st-paragraph says  -- AI wrong prediction
"there is still a high rate of false positives, where predicted binders fail during experimental testing"  ← AI prediction of protein structures is bad.

p.6-upper says  -- CAR-T-cell
"synthetic genes encoding all 44 miBd designs targeting SLLMWITQC/HLA-A*02:01 (= another cancer antigen ) were cloned into a lentiviral CAR vector for expression on the cell surface "

"From this library, we identified one candidate (NY1-B04),"  ← So only one binder-protein out of 44 candidates predicted by AI bound to the target peptide.  ← AI success rate was very bad = only 1/44

p.8-1st-paragraph-lower says  -- Only 2% AI success rate
"We selected 95 miBds for in vitro testing, constituting a success rate of ~2%"  ← Also in design of binders targeting other peptides, the AI success rate was very bad = only 2%.

↑ As shown here, today's AI's success rates of actual proteins' structures were very bad ~2%, because the experimental (static) protein structures in PDB on which AI, Alphafold are trained are different from actual protein structures inside bodies or cells.

Today's AI, experiments cannot cure cancers.

We have to clarify atomic structures of cancer cells by multi-probe atomic force microscopes, which is hampered by unreal quantum mechanics.

↑ To cure cancers, we should stop relying on T cell's immunity that is useless, often blocked by cancers inside bodies.

And we should directly measure atomic structures of cancer cells (= which direct observation of atoms and cells is impossible in today's X-ray and cryo-EM ) by practical multi-probe atomic force microscopes that can design and directly create molecules or effective drugs binding only to cancer cells to prevent them proliferating.

↑ Only atomic force microscopes (with useful multiple probe tips) can directly observe and know the precise atomic structures over (cancer, HIV, auto-immune) cells and tissues, combined with artificial manipulation or chemical bond breaking ( this-p.17 ) separating attached molecules from the target cells gradually.  → Designing and directly creating drug molecules binding only to the target sick (cancer, HIV autoimmune.. ) cells and preventing them from proliferating as effective treatment.

↑ Development of this useful multi-probe atomic force microscopes has been hampered by the current impractical quantum mechanical atomic model.

Feel free to link to this site.