Add a sample_id column to ancestors

Starts to address #604. Also changes the sampledata format

Author: hyanwong

docs/conf.py

@@ -193,6 +193,7 @@
 # Example configuration for intersphinx: refer to the Python standard library.
 intersphinx_mapping = {
     "https://docs.python.org/3/": None,
+    "https://numpy.org/doc/stable/": None,
     "https://tskit.dev/tskit/docs/stable": None,
     "https://tskit.dev/msprime/docs/stable": None,
     "https://numcodecs.readthedocs.io/en/stable/": None,

tests/test_formats.py

@@ -2132,13 +2132,42 @@ def test_insert_proxy_1_sample(self):
                 sample_data, sample_ids=[i]
             )
             assert ancestors.num_ancestors + 1 == ancestors_extra.num_ancestors
-            inserted = -1
+            inserted = -1  # inserted one should always be the last
             self.assert_ancestor_full_span(ancestors_extra, [inserted])
             assert np.array_equal(
                 ancestors_extra.ancestors_haplotype[inserted],
                 sample_data.sites_genotypes[:, i][used_sites],
             )
 
+    def test_insert_proxy_sample_ids(self):
+        ids = [1, 4]
+        sd, _ = self.get_example_data(10, 10, 40)
+        ancestors = tsinfer.generate_ancestors(sd).insert_proxy_samples(
+            sd, sample_ids=ids
+        )
+        inference_sites = np.isin(sd.sites_position[:], ancestors.sites_position[:])
+        anc_sample_ids = ancestors.ancestors_sample_id[:]
+        assert np.sum(anc_sample_ids != tskit.NULL) == len(ids)
+        for sample_id in ids:
+            assert np.sum(anc_sample_ids == sample_id) == 1
+            anc_sample = np.where(anc_sample_ids == sample_id)[0][0]
+            assert ancestors.ancestors_start[anc_sample] == 0
+            assert ancestors.ancestors_end[anc_sample] == ancestors.num_sites
+            assert len(ancestors.ancestors_focal_sites[anc_sample]) == 0
+
+            haplotype = next(sd.haplotypes([sample_id], sites=inference_sites))[1]
+            assert np.all(ancestors.ancestors_haplotype[anc_sample] == haplotype)
+
+    def test_insert_proxy_different_sample_data(self):
+        ids = [1, 4]
+        sd, _ = self.get_example_data(10, 10, 40)
+        ancestors = tsinfer.generate_ancestors(sd)
+        sd_copy, _ = self.get_example_data(10, 10, num_ancestors=40)
+        ancestors_extra = ancestors.insert_proxy_samples(
+            sd_copy, sample_ids=ids, require_same_sample_data=False
+        )
+        assert np.all(ancestors_extra.ancestors_sample_id[:] == tskit.NULL)
+
     def test_insert_proxy_sample_provenance(self):
         sample_data, _ = self.get_example_data(10, 10, 40)
         ancestors = tsinfer.generate_ancestors(sample_data)
@@ -2181,6 +2210,8 @@ def test_insert_proxy_time_historical_samples(self):
         assert np.array_equal(
             ancestors_extra.ancestors_time[-1], historical_sample_time + epsilon
         )
+        assert np.sum(ancestors_extra.ancestors_sample_id[:] != tskit.NULL) == 1
+        assert ancestors_extra.ancestors_sample_id[-1] == 9
 
         # Test 2 proxies, one historical, specified in different ways / orders
         s_ids = np.array([9, 0])
@@ -2202,6 +2233,9 @@ def test_insert_proxy_time_historical_samples(self):
                 ancestors_extra.ancestors_haplotype[-1], G[:, 0][used_sites]
             )
             assert np.array_equal(ancestors_extra.ancestors_time[-1], epsilon)
+            assert np.sum(ancestors_extra.ancestors_sample_id[:] != tskit.NULL) == 2
+            assert ancestors_extra.ancestors_sample_id[-2] == 9
+            assert ancestors_extra.ancestors_sample_id[-1] == 0
 
     def test_insert_proxy_sample_epsilon(self):
         sample_data, _ = self.get_example_data(10, 10, 40)

tsinfer/formats.py

@@ -2021,6 +2021,10 @@ def haplotypes(self, samples=None, sites=None):
             ``None``, return haplotypes for all sample nodes, otherwise this may be a
             numpy array (or array-like) object (converted to dtype=np.int32).
         :param array sites: A numpy array of sites to use.
+
+
+        :return: An iterator returning sucessive instances of (sample_id, haplotype).
+        :rtype: iter(int, numpy.ndarray(dtype=int8))
         """
         if samples is None:
             samples = np.arange(self.num_samples)
@@ -2113,6 +2117,7 @@ class Ancestor:
     time = attr.ib()
     focal_sites = attr.ib()
     haplotype = attr.ib()
+    sample_id = attr.ib()
 
 
 class AncestorData(DataContainer):
@@ -2150,7 +2155,7 @@ class AncestorData(DataContainer):
     """
 
     FORMAT_NAME = "tsinfer-ancestor-data"
-    FORMAT_VERSION = (3, 0)
+    FORMAT_VERSION = (3, 1)
 
     def __init__(self, sample_data, **kwargs):
         super().__init__(**kwargs)
@@ -2209,6 +2214,13 @@ def __init__(self, sample_data, **kwargs):
             dtype="array:i1",
             compressor=self._compressor,
         )
+        self.data.create_dataset(
+            "ancestors/sample_id",
+            shape=(0,),
+            chunks=chunks,
+            compressor=self._compressor,
+            dtype=np.int32,
+        )
 
         self._alloc_ancestor_writer()
 
@@ -2224,6 +2236,7 @@ def _alloc_ancestor_writer(self):
                 "time": self.ancestors_time,
                 "focal_sites": self.ancestors_focal_sites,
                 "haplotype": self.ancestors_haplotype,
+                "sample_id": self.ancestors_sample_id,
             },
             num_threads=self._num_flush_threads,
         )
@@ -2245,6 +2258,7 @@ def __str__(self):
             ("ancestors/time", zarr_summary(self.ancestors_time)),
             ("ancestors/focal_sites", zarr_summary(self.ancestors_focal_sites)),
             ("ancestors/haplotype", zarr_summary(self.ancestors_haplotype)),
+            ("ancestors/sample_id", zarr_summary(self.ancestors_sample_id)),
         ]
         return super().__str__() + self._format_str(values)
 
@@ -2269,6 +2283,9 @@ def data_equal(self, other):
                 self.ancestors_focal_sites[:], other.ancestors_focal_sites[:]
             )
             and np_obj_equal(self.ancestors_haplotype[:], other.ancestors_haplotype[:])
+            and np.array_equal(
+                self.ancestors_sample_id[:], other.ancestors_sample_id[:]
+            )
         )
 
     @property
@@ -2311,6 +2328,10 @@ def ancestors_focal_sites(self):
     def ancestors_haplotype(self):
         return self.data["ancestors/haplotype"]
 
+    @property
+    def ancestors_sample_id(self):
+        return self.data["ancestors/sample_id"]
+
     @property
     def ancestors_length(self):
         """
@@ -2329,6 +2350,7 @@ def insert_proxy_samples(
         *,
         sample_ids=None,
         epsilon=None,
+        map_ancestors=False,
         allow_mutation=False,
         require_same_sample_data=True,
         **kwargs,
@@ -2341,7 +2363,8 @@ def insert_proxy_samples(
 
         A *proxy sample ancestor* is an ancestor based upon a known sample. At
         sites used in the full inference process, the haplotype of this ancestor
-        is identical to that of the sample on which it is based. The time of the
+        is identical to that of the sample on which it is based, and the
+        The time of the
         ancestor is taken to be a fraction ``epsilon`` older than the sample on
         which it is based.
 
@@ -2355,11 +2378,11 @@ def insert_proxy_samples(
 
         .. note::
 
-            The proxy sample ancestors inserted here will correspond to extra nodes
-            in the inferred tree sequence. At sites which are not used in the full
+            The proxy sample ancestors inserted here will end up as extra nodes
+            in the inferred tree sequence, but at sites which are not used in the full
             inference process (e.g. sites unique to a single historical sample),
-            these proxy sample ancestor nodes may have a different genotype from
-            their corresponding sample.
+            it is possible for these proxy sample ancestor nodes to have a different
+            genotype from their corresponding sample.
 
         :param SampleData sample_data: The :class:`.SampleData` instance
             from which to select the samples used to create extra ancestors.
@@ -2394,7 +2417,8 @@ def insert_proxy_samples(
             to ensure that the encoding of alleles in ``sample_data`` matches the
             encoding in the current :class:`AncestorData` instance (i.e. that in the
             original :class:`.SampleData` instance on which the current ancestors
-            are based).
+            are based). Note that in this case, the sample_id is not recorded in the
+            returned object.
         :param \\**kwargs: Further arguments passed to the constructor when creating
             the new :class:`AncestorData` instance which will be returned.
 
@@ -2492,7 +2516,11 @@ def insert_proxy_samples(
                     time=proxy_time,
                     focal_sites=[],
                     haplotype=haplotype,
+                    sample_id=sample_id
+                    if sample_data.uuid == self.sample_data_uuid
+                    else tskit.NULL,
                 )
+
             # Add any ancestors remaining in the current instance
             while ancestor is not None:
                 other.add_ancestor(**attr.asdict(ancestor, filter=exclude_id))
@@ -2574,7 +2602,6 @@ def truncate_ancestors(
         start = self.ancestors_start[:]
         end = self.ancestors_end[:]
         time = self.ancestors_time[:]
-        focal_sites = self.ancestors_focal_sites[:]
         haplotypes = self.ancestors_haplotype[:]
         if upper_time_bound > np.max(time) or lower_time_bound > np.max(time):
             raise ValueError("Time bounds cannot be greater than older ancestor")
@@ -2612,16 +2639,12 @@ def truncate_ancestors(
                     )
                     start[anc.id] = insert_pos_start
                     end[anc.id] = insert_pos_end
-                    time[anc.id] = anc.time
-                    focal_sites[anc.id] = anc.focal_sites
                     haplotypes[anc.id] = anc.haplotype[
                         insert_pos_start - anc.start : insert_pos_end - anc.start
                     ]
                     # TODO - record truncation in ancestors' metadata when supported
         truncated.ancestors_start[:] = start
         truncated.ancestors_end[:] = end
-        truncated.ancestors_time[:] = time
-        truncated.ancestors_focal_sites[:] = focal_sites
         truncated.ancestors_haplotype[:] = haplotypes
         truncated.record_provenance(command="truncate_ancestors")
         truncated.finalise()
@@ -2642,6 +2665,12 @@ def set_inference_sites(self, site_ids):
         sites in the sample data file, and the IDs must be in increasing order.
 
         This must be called before the first call to :meth:`.add_ancestor`.
+
+        .. note::
+            To obtain a list of which sites in a sample data or a tree sequence have
+            been placed into the ancestors file for use in inference, you can apply
+            :func:`numpy.isin` to the list of positions, e.g.
+            ``np.isin(sample_data.sites_position[:], ancestors.sites_position[:])``
         """
         self._check_build_mode()
         position = self.sample_data.sites_position[:][site_ids]
@@ -2650,12 +2679,18 @@ def set_inference_sites(self, site_ids):
         array[:] = position
         self._num_alleles = self.sample_data.num_alleles(site_ids)
 
-    def add_ancestor(self, start, end, time, focal_sites, haplotype):
+    def add_ancestor(
+        self, start, end, time, focal_sites, haplotype, sample_id=tskit.NULL
+    ):
         """
         Adds an ancestor with the specified haplotype, with ancestral material over the
         interval [start:end], that is associated with the specified timepoint and has new
-        mutations at the specified list of focal sites. Ancestors should be added in time
-        order, with the oldest first. The id of the added ancestor is returned.
+        mutations at the specified list of focal sites. If this ancestor is based on a
+        specific sample from the associated sample_data file (i.e. a historical sample)
+        then the ``sample_id`` in the sample data file can also be passed as a parameter.
+
+        The Ancestors should be added in time order, with the oldest first. The id of
+        the added ancestor is returned.
         """
         self._check_build_mode()
         haplotype = tskit.util.safe_np_int_cast(haplotype, dtype=np.int8, copy=True)
@@ -2685,6 +2720,7 @@ def add_ancestor(self, start, end, time, focal_sites, haplotype):
             time=time,
             focal_sites=focal_sites,
             haplotype=haplotype,
+            sample_id=sample_id,
         )
 
     def finalise(self):
@@ -2706,6 +2742,7 @@ def ancestors(self):
         end = self.ancestors_end[:]
         time = self.ancestors_time[:]
         focal_sites = self.ancestors_focal_sites[:]
+        sample_id = self.ancestors_sample_id[:]
         for j, h in enumerate(chunk_iterator(self.ancestors_haplotype)):
             yield Ancestor(
                 id=j,
@@ -2714,6 +2751,7 @@ def ancestors(self):
                 time=time[j],
                 focal_sites=focal_sites[j],
                 haplotype=h,
+                sample_id=sample_id[j],
             )