seqtree 0.1.0__tar.gz → 0.2.0__tar.gz

{seqtree-0.1.0 → seqtree-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: seqtree
-Version: 0.1.0
+Version: 0.2.0
 Summary: Fast fuzzy search over biological sequences (C++ core, Python bindings)
 Keywords: sequence-search,fuzzy-matching,CDR3,immunology,bioinformatics,trie
 Author-Email: ISALGO laboratory <mikhail.shugay@gmail.com>
@@ -59,8 +59,13 @@ to their own payloads (V gene, MHC, counts) and filter.
 Beyond search, seqtree ships:
 
 - **Substitution matrices** — built-in `identity`, `BLOSUM62`, `PAM250`, `PAM100`, and `structural`
-  (TeXshade sidechain volume + hydropathy), plus custom matrices via
-  `SubstitutionMatrix.from_similarity` (Gram-distance penalty `s(a,a)+s(b,b)−2·s(a,b)`).
+  — a **Miyazawa–Jernigan interaction-strength** matrix: each residue's strength `q(a)=mean_b e(a,b)`
+  is read off the MJ contact potential, so substitutions between residues of like interaction strength
+  are cheap. It separates strong (hydrophobic `F W C L Y M I V`) from weak (polar/charged
+  `S Q D E K`) interactors — the strong/weak-interactor axis of TCR-recognition models
+  ([Košmrlj et al., *PNAS* 2008](https://doi.org/10.1073/pnas.0808081105); MJ contact energies from
+  Miyazawa & Jernigan, *J Mol Biol* 1996) — letting dissimilar-but-chemically-equivalent loops align.
+  Plus custom matrices via `SubstitutionMatrix.from_similarity` (Gram penalty `s(a,a)+s(b,b)−2·s(a,b)`).
 - **E-values / significance** — calibrate hit counts against a background control repertoire
   (`load_control` + `evalues`), the TCRNET approach on a finite-sample footing. See the
   [E-value guide](https://antigenomics.github.io/seqtree/evalue.html).

{seqtree-0.1.0 → seqtree-0.2.0}/README.md

@@ -26,8 +26,13 @@ to their own payloads (V gene, MHC, counts) and filter.
 Beyond search, seqtree ships:
 
 - **Substitution matrices** — built-in `identity`, `BLOSUM62`, `PAM250`, `PAM100`, and `structural`
-  (TeXshade sidechain volume + hydropathy), plus custom matrices via
-  `SubstitutionMatrix.from_similarity` (Gram-distance penalty `s(a,a)+s(b,b)−2·s(a,b)`).
+  — a **Miyazawa–Jernigan interaction-strength** matrix: each residue's strength `q(a)=mean_b e(a,b)`
+  is read off the MJ contact potential, so substitutions between residues of like interaction strength
+  are cheap. It separates strong (hydrophobic `F W C L Y M I V`) from weak (polar/charged
+  `S Q D E K`) interactors — the strong/weak-interactor axis of TCR-recognition models
+  ([Košmrlj et al., *PNAS* 2008](https://doi.org/10.1073/pnas.0808081105); MJ contact energies from
+  Miyazawa & Jernigan, *J Mol Biol* 1996) — letting dissimilar-but-chemically-equivalent loops align.
+  Plus custom matrices via `SubstitutionMatrix.from_similarity` (Gram penalty `s(a,a)+s(b,b)−2·s(a,b)`).
 - **E-values / significance** — calibrate hit counts against a background control repertoire
   (`load_control` + `evalues`), the TCRNET approach on a finite-sample footing. See the
   [E-value guide](https://antigenomics.github.io/seqtree/evalue.html).

{seqtree-0.1.0 → seqtree-0.2.0}/include/seqtree/seqtree.hpp

@@ -39,7 +39,7 @@ public:
     static SubstitutionMatrix blosum62();
     static SubstitutionMatrix pam250();   // EMBOSS EPAM250 (NCBI log-odds)
     static SubstitutionMatrix pam100();   // EMBOSS EPAM100 (NCBI log-odds)
-    static SubstitutionMatrix structural();  // TeXshade volume+hydropathy similarity
+    static SubstitutionMatrix structural();  // Miyazawa-Jernigan interaction-strength similarity
     // Convert a similarity matrix (row-major, size*size) to penalties via the Gram /
     // squared-distance transform pen[a][b] = sim[a][a] + sim[b][b] - 2*sim[a][b].
     static SubstitutionMatrix from_similarity(uint8_t size, const int32_t* sim);

{seqtree-0.1.0 → seqtree-0.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "scikit_build_core.build"
 
 [project]
 name = "seqtree"
-version = "0.1.0"
+version = "0.2.0"
 description = "Fast fuzzy search over biological sequences (C++ core, Python bindings)"
 readme = "README.md"
 requires-python = ">=3.10"

seqtree-0.2.0/src/structural.inc

@@ -0,0 +1,31 @@
+// Miyazawa-Jernigan interaction-strength similarity (strong/weak interactors; 0..10).
+// Row/column order: A R N D C Q E G H I L K M F P S T W Y V B Z X *
+// Generated by bench/gen_matrices.py -- do not edit by hand.
+static const int32_t kStructuralSize = 24;
+static const int32_t kStructural[24 * 24] = {
+//    A    R    N    D    C    Q    E    G    H    I    L    K    M    F    P    S    T    W    Y    V    B    Z    X    *
+     10,   9,   9,   7,   6,   8,   7,  10,   9,   6,   6,   5,   6,   5,   9,   8,   9,   6,   6,   7,   8,   7,   7,   0,
+      9,  10,  10,   8,   4,   9,   8,   8,   8,   5,   4,   6,   5,   4,  10,  10,   9,   4,   5,   6,   9,   9,   7,   0,
+      9,  10,  10,   8,   4,   9,   8,   8,   8,   5,   4,   6,   5,   4,  10,  10,   9,   4,   5,   6,   9,   9,   7,   0,
+      7,   8,   8,  10,   3,   9,  10,   7,   6,   3,   3,   8,   3,   2,   8,   9,   8,   3,   3,   4,   9,   9,   6,   0,
+      6,   4,   4,   3,  10,   4,   2,   6,   7,   9,  10,   1,   9,   9,   4,   4,   5,  10,  10,   8,   4,   3,   6,   0,
+      8,   9,   9,   9,   4,  10,   9,   8,   7,   4,   4,   7,   4,   3,   9,  10,   9,   3,   4,   5,   9,   9,   7,   0,
+      7,   8,   8,  10,   2,   9,  10,   6,   6,   3,   2,   8,   3,   2,   8,   8,   7,   2,   3,   4,   9,   9,   6,   0,
+     10,   8,   8,   7,   6,   8,   6,  10,   9,   7,   6,   5,   6,   5,   8,   8,   9,   6,   6,   8,   8,   7,   7,   0,
+      9,   8,   8,   6,   7,   7,   6,   9,  10,   7,   7,   4,   7,   6,   8,   7,   8,   6,   7,   8,   7,   6,   7,   0,
+      6,   5,   5,   3,   9,   4,   3,   7,   7,  10,   9,   1,  10,   9,   5,   5,   6,   9,  10,   9,   4,   4,   7,   0,
+      6,   4,   4,   3,  10,   4,   2,   6,   7,   9,  10,   1,  10,   9,   4,   4,   5,  10,  10,   8,   4,   3,   6,   0,
+      5,   6,   6,   8,   1,   7,   8,   5,   4,   1,   1,  10,   1,   0,   6,   6,   6,   0,   1,   2,   7,   8,   4,   0,
+      6,   5,   5,   3,   9,   4,   3,   6,   7,  10,  10,   1,  10,   9,   5,   5,   5,   9,  10,   9,   4,   4,   7,   0,
+      5,   4,   4,   2,   9,   3,   2,   5,   6,   9,   9,   0,   9,  10,   4,   4,   4,  10,   9,   8,   3,   2,   6,   0,
+      9,  10,  10,   8,   4,   9,   8,   8,   8,   5,   4,   6,   5,   4,  10,  10,   9,   4,   5,   6,   9,   9,   7,   0,
+      8,  10,  10,   9,   4,  10,   8,   8,   7,   5,   4,   6,   5,   4,  10,  10,   9,   4,   4,   6,   9,   9,   7,   0,
+      9,   9,   9,   8,   5,   9,   7,   9,   8,   6,   5,   6,   5,   4,   9,   9,  10,   5,   5,   7,   9,   8,   7,   0,
+      6,   4,   4,   3,  10,   3,   2,   6,   6,   9,  10,   0,   9,  10,   4,   4,   5,  10,   9,   8,   3,   3,   6,   0,
+      6,   5,   5,   3,  10,   4,   3,   6,   7,  10,  10,   1,  10,   9,   5,   4,   5,   9,  10,   9,   4,   3,   6,   0,
+      7,   6,   6,   4,   8,   5,   4,   8,   8,   9,   8,   2,   9,   8,   6,   6,   7,   8,   9,  10,   5,   5,   7,   0,
+      8,   9,   9,   9,   4,   9,   9,   8,   7,   4,   4,   7,   4,   3,   9,   9,   9,   3,   4,   5,   9,   9,   7,   0,
+      7,   9,   9,   9,   3,   9,   9,   7,   6,   4,   3,   8,   4,   2,   9,   9,   8,   3,   3,   5,   9,   9,   6,   0,
+      7,   7,   7,   6,   6,   7,   6,   7,   7,   7,   6,   4,   7,   6,   7,   7,   7,   6,   6,   7,   7,   6,   7,   0,
+      0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  10,
+};

seqtree-0.1.0/src/structural.inc

@@ -1,31 +0,0 @@
-// TeXshade structural similarity (sidechain volume + hydropathy; 0..10).
-// Row/column order: A R N D C Q E G H I L K M F P S T W Y V B Z X *
-// Generated by bench/gen_matrices.py -- do not edit by hand.
-static const int32_t kStructuralSize = 24;
-static const int32_t kStructural[24 * 24] = {
-//    A    R    N    D    C    Q    E    G    H    I    L    K    M    F    P    S    T    W    Y    V    B    Z    X    *
-     10,   3,   5,   6,   3,   5,   6,   8,   3,   3,   3,   5,   5,   5,   8,   8,   8,   3,   3,   8,   6,   6,   5,   0,
-      3,  10,   5,   3,   3,   5,   5,   5,   6,   3,   3,   8,   3,   1,   5,   5,   5,   3,   1,   3,   4,   5,   4,   0,
-      5,   5,  10,   8,   3,   5,   6,   5,   6,   3,   1,   6,   1,   3,   3,   8,   6,   0,   5,   3,   9,   6,   5,   0,
-      6,   3,   8,  10,   1,   6,   8,   6,   5,   1,   1,   5,   3,   1,   5,   6,   5,   0,   3,   5,   9,   7,   4,   0,
-      3,   3,   3,   1,  10,   1,   0,   5,   3,   3,   3,   0,   3,   5,   3,   6,   3,   5,   5,   3,   2,   0,   3,   0,
-      5,   5,   5,   6,   1,  10,   6,   3,   6,   1,   3,   6,   3,   1,   5,   5,   5,   1,   3,   3,   6,   8,   4,   0,
-      6,   5,   6,   8,   0,   6,  10,   6,   3,   1,   1,   6,   3,   3,   5,   5,   5,   1,   1,   6,   7,   8,   4,   0,
-      8,   5,   5,   6,   5,   3,   6,  10,   1,   3,   3,   3,   1,   3,   6,   8,   6,   5,   3,   6,   6,   4,   5,   0,
-      3,   6,   6,   5,   3,   6,   3,   1,  10,   3,   5,   5,   3,   3,   5,   5,   3,   1,   5,   1,   6,   4,   4,   0,
-      3,   3,   3,   1,   3,   1,   1,   3,   3,  10,   8,   3,   6,   6,   3,   3,   5,   5,   5,   3,   2,   1,   4,   0,
-      3,   3,   1,   1,   3,   3,   1,   3,   5,   8,  10,   3,   8,   6,   5,   3,   3,   6,   5,   3,   1,   2,   4,   0,
-      5,   8,   6,   5,   0,   6,   6,   3,   5,   3,   3,  10,   3,   1,   3,   5,   6,   1,   1,   5,   6,   6,   4,   0,
-      5,   3,   1,   3,   3,   3,   3,   1,   3,   6,   8,   3,  10,   5,   3,   4,   5,   5,   3,   6,   2,   3,   4,   0,
-      5,   1,   3,   1,   5,   1,   3,   3,   3,   6,   6,   1,   5,  10,   5,   5,   3,   5,   8,   6,   2,   2,   4,   0,
-      8,   5,   3,   5,   3,   5,   5,   6,   5,   3,   5,   3,   3,   5,  10,   6,   6,   3,   3,   6,   4,   5,   5,   0,
-      8,   5,   8,   6,   6,   5,   5,   8,   5,   3,   3,   5,   4,   5,   6,  10,   8,   3,   5,   6,   7,   5,   6,   0,
-      8,   5,   6,   5,   3,   5,   5,   6,   3,   5,   3,   6,   5,   3,   6,   8,  10,   1,   3,   6,   6,   5,   5,   0,
-      3,   3,   0,   0,   5,   1,   1,   5,   1,   5,   6,   1,   5,   5,   3,   3,   1,  10,   5,   5,   0,   1,   3,   0,
-      3,   1,   5,   3,   5,   3,   1,   3,   5,   5,   5,   1,   3,   8,   3,   5,   3,   5,  10,   5,   4,   2,   4,   0,
-      8,   3,   3,   5,   3,   3,   6,   6,   1,   3,   3,   5,   6,   6,   6,   6,   6,   5,   5,  10,   4,   4,   5,   0,
-      6,   4,   9,   9,   2,   6,   7,   6,   6,   2,   1,   6,   2,   2,   4,   7,   6,   0,   4,   4,   9,   6,   4,   0,
-      6,   5,   6,   7,   0,   8,   8,   4,   4,   1,   2,   6,   3,   2,   5,   5,   5,   1,   2,   4,   6,   8,   4,   0,
-      5,   4,   5,   4,   3,   4,   4,   5,   4,   4,   4,   4,   4,   4,   5,   6,   5,   3,   4,   5,   4,   4,   4,   0,
-      0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,   0,  10,
-};