How a cell shows
what is inside it
To spot an infected cell the immune system must look inside it. But the membrane is opaque.
The answer was a noticeboard โ a cell chops up its own proteins and hangs the fragments on its surface. These 20 lessons follow how that board is built, why it differs between people, and how cancer takes it down.
Not just in words: from 30,000 IMGT/HLA alleles and IEDB mass-spectrometry peptides you draw the polymorphism map yourself, find the binding motifs, build a predictor by hand and score it on a different experiment. Python runs right inside this page.
How you'll learn
Not a simulation. Alleles are counted from the IMGT/HLA file itself; 32,000 mass-spec-confirmed 9mers are pulled from IEDB's 9.2 GB peptide file; mutations come from 10,443 TCGA samples.
In lesson 8 you stack 5,880 HLA-A sequences and count position by position. The positions that come out โ 9, 62, 97, 114, 116, 152 โ are exactly the published binding pockets. Without ever looking at a structure.
Two rulers on the same data give two different pictures. Change the anchor threshold and the anchors change; change the gene list and the 'defect rate' moves tenfold. The moment you fix a criterion you have fixed the answer.
We did not count LOH, did not ship other populations' frequencies, could not adjust for TMB. Lesson 20 splits what this data supports from what it does not in a table โ that habit is the point of the course.