diff --git a/.gitattributes b/.gitattributes
new file mode 100644
index 0000000..2197dc0
--- /dev/null
+++ b/.gitattributes
@@ -0,0 +1 @@
+tests/data/train_smoke/* filter=lfs diff=lfs merge=lfs -text
diff --git a/.github/workflows/docker-build.yml b/.github/workflows/docker-build.yml
index 87f972c..975b38b 100644
--- a/.github/workflows/docker-build.yml
+++ b/.github/workflows/docker-build.yml
@@ -25,9 +25,9 @@ jobs:
 
     steps:
       - name: Checkout repository
-        uses: actions/checkout@v3
+        uses: actions/checkout@v4
         with:
-          fetch-depth: "0"
+          lfs: 'true'
 
       - name: Log in to the Container registry
         uses: docker/login-action@v2
diff --git a/.github/workflows/main.yml b/.github/workflows/main.yml
index 045c425..1aec4f7 100644
--- a/.github/workflows/main.yml
+++ b/.github/workflows/main.yml
@@ -21,7 +21,9 @@ jobs:
   linting:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v1
+      - uses: actions/checkout@v4
+        with:
+          lfs: 'true'
 
       - name: Install Python
         uses: actions/setup-python@v2
@@ -47,7 +49,9 @@ jobs:
   sphinx-docs:
     runs-on: ubuntu-latest
     steps:
-      - uses: actions/checkout@v1
+      - uses: actions/checkout@v4
+        with:
+          lfs: 'true'
 
       - name: Install Python
         uses: actions/setup-python@v2
@@ -76,9 +80,9 @@ jobs:
           - '3.11'
     needs: linting
     steps:
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v4
         with:
-          fetch-depth: 0 # we need tags for versioneer to work
+          lfs: 'true'
 
       - name: Install Python
         uses: actions/setup-python@v2
diff --git a/.github/workflows/release-please.yml b/.github/workflows/release-please.yml
index 4149e3d..84d2519 100644
--- a/.github/workflows/release-please.yml
+++ b/.github/workflows/release-please.yml
@@ -17,10 +17,10 @@ jobs:
           package-name: cada-prio
           token: ${{ secrets.BOT_TOKEN }}
 
-      - uses: actions/checkout@v2
+      - uses: actions/checkout@v4
         if: ${{ steps.release.outputs.release_created }}
         with:
-          fetch-depth: 0
+          lfs: 'true'
 
       - name: Set up Python
         if: ${{ steps.release.outputs.release_created }}
diff --git a/.gitignore b/.gitignore
index 9996ce3..99e20a8 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,3 +1,5 @@
+/model-out
+
 *~
 .*.sw?
 
diff --git a/CONTRIBUTING.rst b/CONTRIBUTING.rst
index d458142..8dbe42c 100644
--- a/CONTRIBUTING.rst
+++ b/CONTRIBUTING.rst
@@ -16,7 +16,7 @@ Types of Contributions
 Report Bugs
 ===========
 
-Report bugs at https://github.com/bihealth/cada-ng/issues.
+Report bugs at https://github.com/bihealth/cada-prio/issues.
 
 If you are reporting a bug, please include:
 
@@ -39,12 +39,12 @@ Anything tagged with "enhancement" and "help wanted" is open to whoever wants to
 Write Documentation
 ===================
 
-cada-ng could always use more documentation, whether as part of the official cada-ng docs, in docstrings, or even on the web in blog posts, articles, and such.
+cada-prio could always use more documentation, whether as part of the official cada-prio docs, in docstrings, or even on the web in blog posts, articles, and such.
 
 Submit Feedback
 ===============
 
-The best way to send feedback is to file an issue at https://github.com/bihealth/cada-ng/issues.
+The best way to send feedback is to file an issue at https://github.com/bihealth/cada-prio/issues.
 
 If you are proposing a feature:
 
@@ -56,18 +56,18 @@ If you are proposing a feature:
 Get Started!
 ------------
 
-Ready to contribute? Here's how to set up `cada-ng` for local development.
+Ready to contribute? Here's how to set up `cada-prio` for local development.
 
-1. Fork the `cada-ng` repo on GitHub.
+1. Fork the `cada-prio` repo on GitHub.
 2. Clone your fork locally::
 
-    $ git clone git@github.com:bihealth/cada-ng.git
+    $ git clone git@github.com:bihealth/cada-prio.git
 
 3. Install your local copy into a virtualenv.
    Assuming you have virtualenvwrapper installed, this is how you set up your fork for local development::
 
-    $ mkvirtualenv cada-ng
-    $ cd cada-ng/
+    $ mkvirtualenv cada-prio
+    $ cd cada-prio/
     $ python setup.py develop
 
 4. Create a branch for local development::
@@ -79,7 +79,7 @@ Ready to contribute? Here's how to set up `cada-ng` for local development.
 5. When you're done making changes, check that your changes pass flake8 and the
    tests, including testing other Python versions with tox::
 
-    $ flake8 cada-ng tests
+    $ flake8 cada-prio tests
     $ python setup.py test or pytest
     $ tox
 
diff --git a/README.md b/README.md
index dde797d..928a71c 100644
--- a/README.md
+++ b/README.md
@@ -1,16 +1,16 @@
-[![CI](https://github.com/bihealth/cada-ng/actions/workflows/main.yml/badge.svg?branch=main)](https://github.com/bihealth/cada-ng/actions/workflows/main.yml)
-[![codecov](https://codecov.io/gh/bihealth/cada-ng/graph/badge.svg?token=HIBwaG4eYM)](https://codecov.io/gh/bihealth/cada-ng)
-[![Documentation Status](https://readthedocs.org/projects/cada-ng/badge/?version=latest)](https://cada-ng.readthedocs.io/en/latest/?badge=latest)
-[![Pypi](https://img.shields.io/pypi/pyversions/cada-ng.svg)](https://pypi.org/project/cada-ng)
+[![CI](https://github.com/bihealth/cada-prio/actions/workflows/main.yml/badge.svg?branch=main)](https://github.com/bihealth/cada-prio/actions/workflows/main.yml)
+[![codecov](https://codecov.io/gh/bihealth/cada-prio/graph/badge.svg?token=HIBwaG4eYM)](https://codecov.io/gh/bihealth/cada-prio)
+[![Documentation Status](https://readthedocs.org/projects/cada-prio/badge/?version=latest)](https://cada-prio.readthedocs.io/en/latest/?badge=latest)
+[![Pypi](https://img.shields.io/pypi/pyversions/cada-prio.svg)](https://pypi.org/project/cada-prio)
 
 # CADA: The Next Generation
 
 This is a re-implementation of the [CADA](https://github.com/Chengyao-Peng/CADA) method for phenotype-similarity prioritization.
 
 - Free software: MIT license
-- Documentation: https://cada-ng.readthedocs.io/en/latest/
-- Discussion Forum: https://github.com/bihealth/cada-ng/discussions
-- Bug Reports: https://github.com/bihealth/cada-ng/issues
+- Documentation: https://cada-prio.readthedocs.io/en/latest/
+- Discussion Forum: https://github.com/bihealth/cada-prio/discussions
+- Bug Reports: https://github.com/bihealth/cada-prio/issues
 
 
 ## Managing GitHub Project with Terraform
@@ -22,7 +22,7 @@ This is a re-implementation of the [CADA](https://github.com/Chengyao-Peng/CADA)
 # cd utils/terraform
 
 # terraform init
-# terraform import github_repository.cada-ng cada-ng
+# terraform import github_repository.cada-prio cada-prio
 # terraform validate
 # terraform fmt
 # terraform plan
diff --git a/cada_prio/cli.py b/cada_prio/cli.py
index 917ceb2..7c36191 100644
--- a/cada_prio/cli.py
+++ b/cada_prio/cli.py
@@ -2,7 +2,7 @@
 
 import click
 
-from cada_prio import _version
+from cada_prio import _version, train_model
 
 
 @click.group()
@@ -13,3 +13,31 @@ def cli(ctx: click.Context, verbose: bool):
     """Main entry point for CLI via click."""
     ctx.ensure_object(dict)
     ctx.obj["verbose"] = verbose
+
+
+@cli.command("train-model")
+@click.argument("path_out", type=str)
+@click.option("--path-hgnc-json", type=str, help="path to HGNC JSON", required=True)
+@click.option(
+    "--path-gene-hpo-links", type=str, help="path to gene-HPO term links JSONL", required=True
+)
+@click.option(
+    "--path-hpo-genes-to-phenotype",
+    type=str,
+    help="path to genes_to_phenotype.txt file",
+    required=True,
+)
+@click.option("--path-hpo-obo", type=str, help="path HPO OBO file", required=True)
+@click.pass_context
+def cli_train_model(
+    ctx: click.Context,
+    path_out: str,
+    path_hgnc_json: str,
+    path_gene_hpo_links: str,
+    path_hpo_genes_to_phenotype: str,
+    path_hpo_obo: str,
+):
+    ctx.ensure_object(dict)
+    train_model.run(
+        path_out, path_hgnc_json, path_gene_hpo_links, path_hpo_genes_to_phenotype, path_hpo_obo
+    )
diff --git a/cada_prio/train_model.py b/cada_prio/train_model.py
new file mode 100644
index 0000000..048aab6
--- /dev/null
+++ b/cada_prio/train_model.py
@@ -0,0 +1,288 @@
+import csv
+import enum
+import itertools
+import json
+import os
+import typing
+import warnings
+
+import attrs
+import cattrs
+from logzero import logger
+import networkx as nx
+import node2vec
+import pronto
+
+
+@attrs.frozen(auto_attribs=True)
+class GeneIds:
+    """Mapping between gene IDs"""
+
+    symbol: str
+    hgnc_id: str
+    ncbi_gene_id: str
+    ensembl_gene_id: str
+
+
+@enum.unique
+class ClinSig(enum.Enum):
+    BENIGN = "benign"
+    LIKELY_BENIGN = "likely benign"
+    UNCERTAIN = "uncertain significance"
+    LIKELY_PATHOGENIC = "likely pathogenic"
+    PATHOGENIC = "pathogenic"
+
+
+@attrs.frozen(auto_attribs=True)
+class GenePhenotypeRecord:
+    """Full record from JSONL file"""
+
+    #: RCV accession
+    rcv: str
+    #: SCV accession
+    scv: str
+    #: Clinical significance
+    clinsig: typing.Optional[ClinSig]
+    #: Submitter
+    submitter: typing.Optional[str]
+    #: Gene HGNC ID
+    hgnc_ids: typing.List[str]
+    #: Linked OMIM terms
+    omim_terms: typing.List[str]
+    #: Linked MONDO terms
+    mondo_terms: typing.List[str]
+    #: Linked HPO terms
+    hpo_terms: typing.List[str]
+
+
+@attrs.frozen(auto_attribs=True, order=True)
+class GeneToPhenotypeRecord:
+    """Minimal link record"""
+
+    #: Submitter
+    submitter: str
+    #: Gene HGNC ID
+    hgnc_id: str
+    #: Linked HPO terms
+    hpo_terms: typing.Tuple[str, ...]
+
+
+def load_gene_phenotype_records(path: str) -> typing.Iterable[GenePhenotypeRecord]:
+    with open(path) as f:
+        for line in f:
+            yield cattrs.structure(json.loads(line), GenePhenotypeRecord)
+
+
+def load_hpo_ontology(
+    path_hpo_obo,
+) -> typing.Tuple[pronto.Ontology, typing.Dict[str, str], typing.Dict[str, str]]:
+    logger.info("Loading HPO OBO from %s...", path_hpo_obo)
+    logger.info("- parsing ...")
+    with warnings.catch_warnings():
+        warnings.filterwarnings("ignore", category=UnicodeWarning)
+        hpo_ontology = pronto.Ontology(path_hpo_obo)
+    logger.info("- extracting term ids ...")
+    hpo_id_from_alt: typing.Dict[str, str] = {}
+    hpo_id_to_name: typing.Dict[str, str] = {}
+    for term in hpo_ontology.terms():
+        if not term.name or not term.id:
+            continue
+        hpo_id_to_name[term.id] = term.name
+        for alternate_id in term.alternate_ids:
+            hpo_id_from_alt[alternate_id] = term.id
+            hpo_id_to_name[alternate_id] = term.name
+    logger.info("... done loading %s terms", len(hpo_ontology))
+
+    return hpo_ontology, hpo_id_from_alt, hpo_id_to_name
+
+
+def load_hpo_gen2phen(
+    path_hpo_genes_to_phenotype: str, ncbi_to_hgnc: typing.Dict[str, str]
+) -> typing.List[GeneToPhenotypeRecord]:
+    logger.info("Loading HPO gene-phenotype links from %s...", path_hpo_genes_to_phenotype)
+    logger.info("- parsing records ...")
+    warned_about = set()
+    with open(path_hpo_genes_to_phenotype, "rt") as inputf:
+        reader = csv.DictReader(inputf, delimiter="\t")
+        hpo_by_gene: typing.Dict[str, typing.Set[str]] = {}
+        for record in reader:
+            ncbi_gene_id = record["ncbi_gene_id"]
+            if ncbi_gene_id not in ncbi_to_hgnc:
+                if ncbi_gene_id not in warned_about:
+                    logger.warning("No HGNC entry for Entrez %s found", ncbi_gene_id)
+                    warned_about.add(ncbi_gene_id)
+                continue
+            hgnc_id = ncbi_to_hgnc[ncbi_gene_id]
+            hpo_by_gene.setdefault(hgnc_id, set()).add(record["hpo_id"])
+
+        hpo_gene_to_phenotype_set = {
+            GeneToPhenotypeRecord(
+                submitter="HPO",
+                hgnc_id=hgnc_id,
+                hpo_terms=tuple(hpo_terms),
+            )
+            for hgnc_id, hpo_terms in hpo_by_gene.items()
+        }
+    logger.info("- sorting ...")
+    hpo_gene_to_phenotype = list(sorted(hpo_gene_to_phenotype_set))
+    logger.info("... done loading %s records", len(hpo_gene_to_phenotype))
+
+    return hpo_gene_to_phenotype
+
+
+def load_clinvar_gen2phen(path_gene_hpo_links: str) -> typing.List[GeneToPhenotypeRecord]:
+    logger.info("Loading ClinVar gene-phenotype links from %s...", path_gene_hpo_links)
+    logger.info("- parsing records ...")
+    clinvar_gene_to_phenotype_set = {
+        GeneToPhenotypeRecord(
+            submitter=record.submitter,
+            hgnc_id=record.hgnc_ids[0],
+            hpo_terms=tuple(record.hpo_terms),
+        )
+        for record in load_gene_phenotype_records(path_gene_hpo_links)
+        if record.clinsig in (ClinSig.LIKELY_PATHOGENIC, ClinSig.PATHOGENIC)
+        and len(record.hgnc_ids) == 1
+        and record.submitter
+    }
+    logger.info("- sorting ...")
+    clinvar_gene_to_phenotype = list(sorted(clinvar_gene_to_phenotype_set))
+    logger.info("... done loading %s records", len(clinvar_gene_to_phenotype_set))
+
+    return clinvar_gene_to_phenotype
+
+
+def load_hgnc_info(path_hgnc_json) -> typing.Tuple[typing.Dict[str, str], typing.List[GeneIds]]:
+    logger.info("Loading HGNC info from %s...", path_hgnc_json)
+    gene_ids = []
+    ncbi_to_hgnc = {}
+    with open(path_hgnc_json) as f:
+        full_hgnc = json.load(f)
+        for doc in full_hgnc["response"]["docs"]:
+            if all(key in doc for key in ("symbol", "entrez_id")):
+                gene_id = GeneIds(
+                    symbol=doc["symbol"],
+                    hgnc_id=doc["hgnc_id"],
+                    ncbi_gene_id=doc["entrez_id"],
+                    ensembl_gene_id=doc.get("ensembl_gene_id"),
+                )
+                ncbi_to_hgnc[gene_id.ncbi_gene_id] = gene_id.hgnc_id
+                gene_ids.append(gene_id)
+    logger.info("... done loading %s records", len(gene_ids))
+
+    return ncbi_to_hgnc, gene_ids
+
+
+#: type alias for an edge.
+Edge = typing.Tuple[str, str]
+
+
+def yield_hpo_edges(hpo_ontology: pronto.Ontology) -> typing.Iterator[Edge]:
+    """Yield hierarchical edges from the HPO ontology"""
+    for term in hpo_ontology.terms():
+        parents = list(term.superclasses(distance=1))
+        for parent in parents[1:]:
+            yield (term.id, parent.id)
+
+
+def yield_gene2phen_edges(links: typing.List[GeneToPhenotypeRecord]) -> typing.Iterator[Edge]:
+    """Yield edges from link record list"""
+    for link in links:
+        for hpo_term in link.hpo_terms:
+            yield (link.hgnc_id, hpo_term)
+
+
+@attrs.frozen(auto_attribs=True)
+class EmbeddingParams:
+    """Parameters for the embedding"""
+
+    #: The number of dimensions of feature vectors
+    dimensions: int = 300
+    #: The number of nodes in each random walk
+    walk_length: int = 60
+    #: Controls the probability for a walk to visit immediately back to the previous node
+    p: float = 1.7987535798694703
+    #: Controls the probability for a walk to visit previously unexplored neighborhoods in the
+    #: graph
+    q: float = 3.875406134463754
+    #: Number of random walks to be generated from each node in the graph
+    num_walks: int = 10
+    #: Limit on the number of words in each context
+    window: int = 4
+    #: Set the min_count in the fitting
+    min_count: int = 1
+    #: Set the batch_words in the fitting
+    batch_words: int = 4
+    #: Number of workers threads to use
+    workers: int = 4
+
+
+def build_and_fit_model(clinvar_gen2phen, hpo_ontology):
+    # create graph edges combining HPO hierarchy and training edges from ClinVar
+    logger.info("Constructing training graph ...")
+    logger.info("- building edges ...")
+    training_edges = list(
+        itertools.chain(yield_hpo_edges(hpo_ontology), yield_gene2phen_edges(clinvar_gen2phen))
+    )
+    logger.info("- graph construction")
+    training_graph = nx.Graph()
+    training_graph.add_edges_from(training_edges)
+    logger.info("... done constructing training graph with %s edges", len(training_edges))
+
+    logger.info("Computing the embedding / model fit...")
+    logger.info("- embedding graph")
+    embedding_params = EmbeddingParams()
+    embedding = node2vec.Node2Vec(
+        training_graph,
+        dimensions=embedding_params.dimensions,
+        walk_length=embedding_params.walk_length,
+        num_walks=embedding_params.num_walks,
+        p=embedding_params.p,
+        q=embedding_params.q,
+        workers=embedding_params.workers,
+    )
+    logger.info("- fitting model")
+    model = embedding.fit(
+        window=embedding_params.window,
+        min_count=embedding_params.min_count,
+        batch_words=embedding_params.batch_words,
+    )
+    logger.info("... done computing the embedding")
+    return training_graph, model
+
+
+def write_graph_and_model(path_out, training_graph, model):
+    os.makedirs(path_out, exist_ok=True)
+
+    path_graph = os.path.join(path_out, "graph.gpickle")
+    logger.info("Saving graph to %s...", path_graph)
+    nx.write_gpickle(training_graph, path_graph)
+    logger.info("... done saving graph")
+
+    logger.info("Saving embedding to %s...", path_out)
+    path_embeddings = os.path.join(path_out, "embedding")
+    logger.info("- %s", path_embeddings)
+    model.wv.save_word2vec_format(path_embeddings)
+    path_model = os.path.join(path_out, "model")
+    logger.info("- %s", path_model)
+    model.save(path_model)
+    logger.info("... done saving embedding to")
+
+
+def run(
+    path_out: str,
+    path_hgnc_json: str,
+    path_gene_hpo_links: str,
+    path_hpo_genes_to_phenotype: str,
+    path_hpo_obo: str,
+):
+    # load all data
+    ncbi_to_hgnc, hgnc_info = load_hgnc_info(path_hgnc_json)
+    clinvar_gen2phen = load_clinvar_gen2phen(path_gene_hpo_links)
+    hpo_gen2phen = load_hpo_gen2phen(path_hpo_genes_to_phenotype, ncbi_to_hgnc)
+    hpo_ontology, hpo_id_from_alt, hpo_id_to_name = load_hpo_ontology(path_hpo_obo)
+    _, _, _, _ = hgnc_info, hpo_gen2phen, hpo_id_from_alt, hpo_id_to_name
+
+    # build and fit model
+    training_graph, model = build_and_fit_model(clinvar_gen2phen, hpo_ontology)
+    # write out graph and model
+    write_graph_and_model(path_out, training_graph, model)
diff --git a/docs/installation.rst b/docs/installation.rst
index 9bf5a51..c42d6bf 100644
--- a/docs/installation.rst
+++ b/docs/installation.rst
@@ -24,10 +24,10 @@ If you don't have `pip <https://pip.pypa.io>`__ installed, this `Python installa
 From Sources
 ------------
 
-The sources for CADA can be downloaded from the `Github repo <https://github.com/bihealth/cada-ng>`__.
+The sources for CADA can be downloaded from the `Github repo <https://github.com/bihealth/cada-prio>`__.
 
 Clone the public repository:
 
 .. code-block:: console
 
-    $ git clone https://github.com/bihealth/cada-ng.git
+    $ git clone https://github.com/bihealth/cada-prio.git
diff --git a/requirements/base.txt b/requirements/base.txt
index c102d56..229cd9d 100644
--- a/requirements/base.txt
+++ b/requirements/base.txt
@@ -4,3 +4,6 @@ cattrs >=22.2.0, <24.0
 click >=8.1.3, <9.0
 toml >=0.10.2, <0.11
 tqdm >=4.0
+pronto >=2.5, <3.0
+networkx
+node2vec >=0.4.6, <0.5
diff --git a/requirements/test.txt b/requirements/test.txt
index 1b36e72..9f9023e 100644
--- a/requirements/test.txt
+++ b/requirements/test.txt
@@ -16,6 +16,7 @@ types-tabulate >=0.9.0.0
 types-jsonschema >=4.17.0
 types-tqdm >=4.66.0
 types-xmltodict >=0.13.0.3
+networkx-stubs
 
 sphinx
 sphinx_rtd_theme
diff --git a/setup.py b/setup.py
index 5123b50..d32c6e9 100644
--- a/setup.py
+++ b/setup.py
@@ -69,7 +69,7 @@ def parse_requirements(path):
     ),
     test_suite="tests",
     tests_require=test_requirements,
-    url="https://github.com/bihealth/cada-ng",
+    url="https://github.com/bihealth/cada-prio",
     version=version,
     zip_safe=False,
 )
diff --git a/stubs/gensim/models.pyi b/stubs/gensim/models.pyi
new file mode 100644
index 0000000..88d53c3
--- /dev/null
+++ b/stubs/gensim/models.pyi
@@ -0,0 +1,7 @@
+class KeyedVectors:
+    def save_word2vec_format(self, fname: str): ...
+
+class Word2Vec:
+    wv: KeyedVectors
+
+    def save(self, fname: str): ...
diff --git a/stubs/node2vec.pyi b/stubs/node2vec.pyi
new file mode 100644
index 0000000..e32561b
--- /dev/null
+++ b/stubs/node2vec.pyi
@@ -0,0 +1,24 @@
+import typing
+
+import gensim.models
+import networkx as nx
+
+class Node2Vec:
+    def __init__(
+        self,
+        graph: nx.Graph,
+        dimensions: int = 128,
+        walk_length: int = 80,
+        num_walks: int = 10,
+        p: float = 1,
+        q: float = 1,
+        weight_key: str = "weight",
+        workers: int = 1,
+        sampling_strategy: typing.Optional[dict] = None,
+        quiet: bool = False,
+        temp_folder: typing.Optional[str] = None,
+        seed: typing.Optional[int] = None,
+    ): ...
+    def fit(
+        self, window: int, min_count: int, batch_words: int, **kwargs
+    ) -> gensim.models.Word2Vec: ...
diff --git a/tests/data/train_smoke/genes_to_phenotype.txt b/tests/data/train_smoke/genes_to_phenotype.txt
new file mode 100644
index 0000000..c03da33
--- /dev/null
+++ b/tests/data/train_smoke/genes_to_phenotype.txt
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d17d73f47b10d9b7741140857341119ec7cb88dced051ea64a674d01bae81dd4
+size 18594201
diff --git a/tests/data/train_smoke/hgnc_complete_set_2023-09-01.json b/tests/data/train_smoke/hgnc_complete_set_2023-09-01.json
new file mode 100644
index 0000000..0560993
--- /dev/null
+++ b/tests/data/train_smoke/hgnc_complete_set_2023-09-01.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e56a3413f01ce7ba9b2d60f4765f4064ee4b8904ed850b7855f0503d2a7d4096
+size 34851645
diff --git a/tests/data/train_smoke/hp.obo b/tests/data/train_smoke/hp.obo
new file mode 100644
index 0000000..6edaa1a
--- /dev/null
+++ b/tests/data/train_smoke/hp.obo
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04ece06aa74a17a74c1be2651a81f2201da88c0d19718f6302db703f8d8651b1
+size 9364837
diff --git a/tests/data/train_smoke/links.jsonl b/tests/data/train_smoke/links.jsonl
new file mode 100644
index 0000000..cfed511
--- /dev/null
+++ b/tests/data/train_smoke/links.jsonl
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:045c4faee127a8cde81e8d25251a5595530cad2b6cb28122624b2deea6d1f164
+size 69157
diff --git a/tests/test_dummy.py b/tests/test_dummy.py
deleted file mode 100644
index f4f5361..0000000
--- a/tests/test_dummy.py
+++ /dev/null
@@ -1,2 +0,0 @@
-def test_dummy():
-    assert True
diff --git a/tests/test_train_model.py b/tests/test_train_model.py
new file mode 100644
index 0000000..18e3181
--- /dev/null
+++ b/tests/test_train_model.py
@@ -0,0 +1,11 @@
+from cada_prio import train_model
+
+
+def test_train_smoke_test(tmpdir):
+    train_model.run(
+        f"{tmpdir}/train_out",
+        "tests/data/train_smoke/hgnc_complete_set_2023-09-01.json",
+        "tests/data/train_smoke/links.jsonl",
+        "tests/data/train_smoke/genes_to_phenotype.txt",
+        "tests/data/train_smoke/hp.obo",
+    )
diff --git a/utils/docker/build-docker.sh b/utils/docker/build-docker.sh
index 75b7cc5..5d26886 100755
--- a/utils/docker/build-docker.sh
+++ b/utils/docker/build-docker.sh
@@ -6,7 +6,7 @@ set -x
 set -euo pipefail
 
 ORG=bihealth
-REPO=cada-ng
+REPO=cada-prio
 DOCKER_VERSION=${DOCKER_VERSION-adhoc}
 
 sudo docker build . \
diff --git a/utils/terraform/main.tf b/utils/terraform/main.tf
index f40eeb9..a36f677 100644
--- a/utils/terraform/main.tf
+++ b/utils/terraform/main.tf
@@ -1,7 +1,7 @@
 # Mangement of the GitHub project.
 
-resource "github_repository" "cada-ng" {
-  name        = "cada-ng"
+resource "github_repository" "cada-prio" {
+  name        = "cada-prio"
   description = "Re-implementation of the CADA phenotype-based prioritization algorithm"
 
   has_issues = true