Initial commit: Chai-1 protein structure prediction pipeline for WES

- Nextflow pipeline using chai1 Docker image from Harbor
- S3-based input/output paths (s3://omic/eureka/chai-lab/)
- GPU-accelerated protein folding with MSA support

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

tests/test_cif_utils.py

@@ -0,0 +1,24 @@
+# Copyright (c) 2024 Chai Discovery, Inc.
+# Licensed under the Apache License, Version 2.0.
+# See the LICENSE file for details.
+
+
+import pytest
+
+from chai_lab.data.io.cif_utils import get_chain_letter
+
+
+def test_get_chain_letter():
+    with pytest.raises(AssertionError):
+        get_chain_letter(0)
+    assert get_chain_letter(1) == "A"
+    assert get_chain_letter(26) == "Z"
+    assert get_chain_letter(27) == "a"
+    assert get_chain_letter(52) == "z"
+
+    assert get_chain_letter(53) == "AA"
+    assert get_chain_letter(54) == "AB"
+
+    # For one-letter codes, there are 26 + 26 = 52 codes
+    # For two-letter codes, there are 52 * 52 codes
+    assert get_chain_letter(52 * 52 + 52) == "zz"