xengsort 2.0.7__tar.gz → 2.0.8__tar.gz

{xengsort-2.0.7 → xengsort-2.0.8}/MANIFEST.in

@@ -2,4 +2,4 @@ include xengsort/xengsort/*
 include xengsort/io/*
 include xengsort/lowlevel/*
 include xengsort/values/*
-include xengsort/config/*
+include xengsort/xengsort/config/*

{xengsort-2.0.7/xengsort.egg-info → xengsort-2.0.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: xengsort
-Version: 2.0.7
+Version: 2.0.8
 Summary: xenograft sorting of FASTQ reads into five categories: host, graft, both, neither, ambiguous.
 Author-email: Jens Zentgraf <zentgraf@cs.uni-saarland.de>, Sven Rahmann <rahmann@cs.uni-saarland.de>
 License: MIT License
@@ -38,8 +38,8 @@ Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
 Requires-Python: >=3.12
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: numpy>=2
-Requires-Dist: numba>=0.60
+Requires-Dist: numpy
+Requires-Dist: numba
 Requires-Dist: pytest
 Requires-Dist: jsonargparse>=4.29.0
 Requires-Dist: pip

{xengsort-2.0.7 → xengsort-2.0.8}/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "xengsort"
-version = "2.0.7"
+version = "2.0.8"
 
 license = {file = "LICENSE"}
 
@@ -21,8 +21,8 @@ readme = "README.md"
 requires-python = ">=3.12"
 
 dependencies = [
-  "numpy>=2",
-  "numba>=0.60",
+  "numpy",
+  "numba",
   "pytest",
   "jsonargparse>=4.29.0",
   "pip",

{xengsort-2.0.7 → xengsort-2.0.8}/xengsort/io/hashio.py

@@ -126,14 +126,14 @@ def save_hash(outname, h, valueinfo, optinfo=dict(), appinfo=dict(), *, _pairs=_
         outname = outname.rsplit(".", 1)[0]
     timestamp0, timestamp1, timestamp2 = debug.timestamp
     debugprint0, debugprint1, debugprint2 = debug.debugprint
-    startout = timestamp0(msg="\n## Output")
     if outname.casefold() in ("/dev/null", "null", "none"):
         debugprint0(f"- not writing special null output file '{outname}'")
         return None
+    startout = timestamp0(msg="- saving hash table '{outname}'")
     debugprint0(f"- writing output files '{outname}.hash', '{outname}.info'...")
     hashinfo = {name: getattr(h, hname) for (name, hname) in _pairs}
     obj = (hashinfo, valueinfo, optinfo, appinfo)
-    pickle.dumps(obj)  # test whether this works before writing huge hash table
+    pickle.dumps(obj)  # dummy dump to string: test if it works before writing huge array
     checksum = write_array(f"{outname}.hash", h.hashtable)
     hashinfo['checksum'] = checksum
     with open(f"{outname}.info", "wb") as fout:
@@ -150,8 +150,10 @@ def load_hash(filename, *, shared_memory=False, info_only=False):
         filename = filename.rsplit(".", 1)[0]
 
     if not info_only:
-        startload = timestamp0(msg="\n## Loading hash table")
+        startload = timestamp0(msg=f"- loading hash table {filename}")
         debugprint0(f"- hash files '{filename}.info', '{filename}.hash'...")
+    else:
+        debugprint1(f"- loading hash info '{filename}.info'")
 
     with open(f"{filename}.info", "rb") as finfo:
         infotup = pickle.load(finfo)
@@ -159,7 +161,7 @@ def load_hash(filename, *, shared_memory=False, info_only=False):
         return None, None, infotup
     (hashinfo, valueinfo, optinfo, appinfo) = infotup
 
-    debugprint1(f"- Importing value set {valueinfo}.")
+    debugprint1(f"- importing value set {valueinfo}.")
     vmodule = import_module(f"..values.{valueinfo[0]}", __package__)
     values = vmodule.initialize(*(valueinfo[1:]))
     update_value = values.update
@@ -183,9 +185,9 @@ def load_hash(filename, *, shared_memory=False, info_only=False):
         debugprint0("- WARNING: Converting hashfuncs from bytes to str")
     checksum = hashinfo['checksum']
 
-    debugprint1(f"- Hash functions: {hashfuncs}")
+    debugprint1(f"- hash functions: {hashfuncs}")
     maxwalk = optinfo.get('maxwalk', -1)
-    debugprint1(f"- Building hash table of type '{hashtype}'...")
+    debugprint1(f"- building hash table of type '{hashtype}'...")
     m = import_module(f"..hash_{hashtype}", __package__)
 
     init = True
@@ -197,26 +199,22 @@ def load_hash(filename, *, shared_memory=False, info_only=False):
         shm_buf = shm.buf
         assert shm_buf.shape[0] % 8 == 0
         assert shm_buf.itemsize == 1
-
         init = np.ndarray(shm_buf.shape[0] // 8, dtype=np.uint64, buffer=shm_buf)
-
     h = m.build_hash(universe, n, subtables,
         bucketsize, hashfuncs, nvalues, update_value,
         aligned=aligned, nfingerprints=nfingerprints,
         init=init, maxwalk=maxwalk, shortcutbits=shortcutbits, shm=shm)
-
     if not shared_memory:
+        debugprint1(f"- loading into '{filename}.hash' into array")
         checked = load_array_into(f"{filename}.hash", h.hashtable, check=checksum)
-
         if checked:
-            debugprint2(f"- Checksum {checksum} successfully verified. Nice.")
-    else:
+            debugprint2(f"- checksum {checksum} successfully verified. Nice.")
+    else:  # checksum for shared memory load
         check = h.hashtable[:256].sum()
         if check != checksum:
-            raise RuntimeError(f"- ERROR load shared memory: {checksum=} does not match expected {check}")
+            raise RuntimeError(f"- ERROR loading into shared memory: {checksum=} does not match expected {check}")
         else:
-            debugprint2(f"- Checksum {checksum} successfully verified. Nice.")
-
+            debugprint2(f"- checksum {checksum} successfully verified. Nice.")
     timestamp0(startload, msg=f"- time to load '{filename}'")
     return h, values, (hashinfo, valueinfo, optinfo, appinfo)
 
@@ -240,7 +238,7 @@ def compile_textdumper(h, f, nvalues, infotup):
     nsubtables, nbuckets, bucketsize = h.subtables, h.nbuckets, h.bucketsize
     universe, choices = h.universe, h.choices
     bits = (bitsfor(universe), bitsfor(choices), bitsfor(nvalues))
-    debugprint1(f"- Bits for (keys, choices, values): {bits}")
+    debugprint1(f"- bits for (keys, choices, values): {bits}")
     powbits = tuple(nextpow2(b) for b in bits)
     NBK, NBC, NBV = [_nbytes[pb] for pb in powbits]
     _keytype, _choicetype, _valuetype = [_nbdtypes[pb] for pb in powbits]
@@ -282,7 +280,7 @@ def compile_dumper(h, f, nvalues, infotup, packed):
     nsubtables, nbuckets, bucketsize = h.subtables, h.nbuckets, h.bucketsize
     universe, choices = h.universe, h.choices
     bits = (bitsfor(universe), bitsfor(choices), bitsfor(nvalues))
-    debugprint1(f"- Bits for (keys, choices, values): {bits}")
+    debugprint1(f"- bits for (keys, choices, values): {bits}")
     # define buffers
     N = 2**14  # N: buffer size in items: 16K or 2**14
     if packed:

xengsort-2.0.8/xengsort/xengsort/config/classify.yaml

@@ -0,0 +1,60 @@
+threads: 8
+prefetchlevel: 0
+chunksize: 8.0
+classification: count
+params_count:
+    nothing: 0
+    few: 6
+    Ag: 3
+    Ah: 3
+    Mh_factor: 4
+    Mh_min: 3
+    Mg_factor: 4
+    Mg_min: 3
+    Mb_factor: 5
+    Mb_min: 3
+    Mn_factor: 4
+    Mn_min: 3
+
+params_coverage:
+    min_size: 1 # k+minsize
+    weak_scale: "5/6" # 0.84 # 5/6
+
+    # at least T1_min host/graft, less than T1_max graft/host and at least T1_strong strong host/graft k-mers
+    T1_min: 0.25
+    T1_max: 0.01
+    T1_strong: 2
+
+    # More than T2 bases are covered by both
+    T2: 0.9
+
+    # More than T3 bases are covered by neither and the score_neither >= T3_score
+    T3: 0.9
+    T3_score: 0.5
+
+    # more than T4 host and graft -> ambiguous
+    T4: 0.26
+
+    # first >= T5 and gap >= T5_gap
+    # T5: 0.65
+    # T5_gap: 0.35
+    T5_both_h_g: 0.1
+    T5: 0.5
+    T5_gap: 0.25
+    T5_both_h_g: 0.1
+
+    # first is host/graft and second is atleast T6 both
+    # T6: 0.6
+    T6: 0.4
+
+    # first is both; second is not neither
+    # T7: 0.5
+    # T7_h_g: 0.21
+    T7: 0.3
+    T7_h_g: 0.1
+
+    # first neither
+    T8: 0.3
+
+    # T9_gap: 0.25
+    T9_gap: 0

xengsort-2.0.8/xengsort/xengsort/config/index.yaml

@@ -0,0 +1,9 @@
+bucketsize: 4
+fill: 0.85
+shortcutbits: 0
+weakthreads: 8
+maxwalk: 500
+maxfailures: 0
+walkseed: 42
+groupprefixlength: 2
+# chunksize: 8.0

{xengsort-2.0.7 → xengsort-2.0.8/xengsort.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: xengsort
-Version: 2.0.7
+Version: 2.0.8
 Summary: xenograft sorting of FASTQ reads into five categories: host, graft, both, neither, ambiguous.
 Author-email: Jens Zentgraf <zentgraf@cs.uni-saarland.de>, Sven Rahmann <rahmann@cs.uni-saarland.de>
 License: MIT License
@@ -38,8 +38,8 @@ Classifier: Topic :: Scientific/Engineering :: Bio-Informatics
 Requires-Python: >=3.12
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: numpy>=2
-Requires-Dist: numba>=0.60
+Requires-Dist: numpy
+Requires-Dist: numba
 Requires-Dist: pytest
 Requires-Dist: jsonargparse>=4.29.0
 Requires-Dist: pip

{xengsort-2.0.7 → xengsort-2.0.8}/xengsort.egg-info/SOURCES.txt

@@ -47,4 +47,6 @@ xengsort/xengsort/xengsort_info.py
 xengsort/xengsort/xengsort_load.py
 xengsort/xengsort/xengsort_main.py
 xengsort/xengsort/xengsort_remove.py
-xengsort/xengsort/xengsort_weak.py
+xengsort/xengsort/xengsort_weak.py
+xengsort/xengsort/config/classify.yaml
+xengsort/xengsort/config/index.yaml

{xengsort-2.0.7 → xengsort-2.0.8}/xengsort.egg-info/requires.txt

@@ -1,5 +1,5 @@
-numpy>=2
-numba>=0.60
+numpy
+numba
 pytest
 jsonargparse>=4.29.0
 pip