From ce4c4492048520395bf444acf83aed79908d6e9a Mon Sep 17 00:00:00 2001
From: Benoit Putzeys <benoit@helical-ai.com>
Date: Tue, 4 Nov 2025 17:31:44 +0100
Subject: [PATCH 1/2] Use list of gene symbols and set nan values to None as
 before

---
 .pybiomart.sqlite                               | Bin 0 -> 24576 bytes
 .../test_transcriptformer_model.py              |   4 ++--
 helical/utils/mapping.py                        |   4 ++--
 3 files changed, 4 insertions(+), 4 deletions(-)
 create mode 100644 .pybiomart.sqlite

diff --git a/.pybiomart.sqlite b/.pybiomart.sqlite
new file mode 100644
index 0000000000000000000000000000000000000000..56cd8403fe16d23218353ad3b4ffb8fe39b3be59
GIT binary patch
literal 24576
zcmeI&Jx{_w7{Ku>ZyFN{abe^V7f6f=ac~v%Vu%Qe4S|VNIg>^Nq{ZOq>i2T<1G#CB
zltxW-akBYe((+vIde`UITQ2nSqU8lz-c09rZXowXQCOBZlu`(h4v$=Ttc$5<)|K$r
z`V{{Zrp4a#cE0c?vUX1tI)xYe-R{}1KXvCi0tg_000IagfB*srAb`OC5(qLIrE=MN
zItbjWv7YOZH`mvJ9|zf^uBtmqI`u|N$@sFY$gte%N9m}(BhS0dvwC+RPt{=8G(5QD
zg_ez0r?IL^y_|*P`m)(})QRd=YuQ{$l`Y{-MtbSrkHd}a&|L)6Rec!GGOWd6uAR=6
zcFLBw>Jp7N^hV2Y*x7XAYd;e`X|sK-`s*1@57}vl4<7ZZM7LH0>ucgsd)zVgE%B)R
zuZgFAPdu8%Bp!ukFTZKlRNkD5=0rgN0R#|0009ILKmY**5I_KdqzY`A#Mu8Q_4Se$
z0tg_000IagfB*srAb<b@D*^WZECd7)KmY**5I_I{1Q0*~0R)mS!1w>jf5&8q00Iag
nfB*srAb<b@2q1s}|NpZF5I_I{1Q0*~0R#|0009ILNWQ=ae|VD;

literal 0
HcmV?d00001

diff --git a/ci/tests/test_transcriptformer/test_transcriptformer_model.py b/ci/tests/test_transcriptformer/test_transcriptformer_model.py
index 60c5adb5..900088c0 100644
--- a/ci/tests/test_transcriptformer/test_transcriptformer_model.py
+++ b/ci/tests/test_transcriptformer/test_transcriptformer_model.py
@@ -22,11 +22,11 @@ def test_process_data__correct_ensembl_ids(self):
         dataset = self.transcriptformer.process_data([self.data])
         assert len(dataset) == 1
         assert all(
-            dataset.files_list[0].var["ensembl_id"].astype(str).values
+            dataset.files_list[0].var["ensembl_id"].values
             == [
                 "ENSG00000121410",
                 "ENSG00000036549",
-                "nan",
+                None,
                 "ENSG00000074755",
             ]
         )
diff --git a/helical/utils/mapping.py b/helical/utils/mapping.py
index 605cd1c6..8ac6daa8 100644
--- a/helical/utils/mapping.py
+++ b/helical/utils/mapping.py
@@ -105,7 +105,7 @@ def convert_list_ensembl_ids_to_gene_symbols(
     """
     df = _get_ensembl_mart_df(species=species)
     mapping = df.drop_duplicates(subset="ensembl_id").set_index("ensembl_id")["gene_name"]
-    return list(pd.Series(ensembl_ids, dtype="object").map(mapping))
+    return list(pd.Series(list(ensembl_ids), dtype="object").map(mapping).where(pd.notna, None))
 
 
 def convert_list_gene_symbols_to_ensembl_ids(
@@ -128,4 +128,4 @@ def convert_list_gene_symbols_to_ensembl_ids(
     """
     df = _get_ensembl_mart_df(species=species)
     mapping = df.drop_duplicates(subset="gene_name").set_index("gene_name")["ensembl_id"]
-    return list(pd.Series(gene_symbols, dtype="object").map(mapping))
+    return list(pd.Series(list(gene_symbols), dtype="object").map(mapping).where(pd.notna, None))
\ No newline at end of file

From be2323c7e968e468f9e0c3afffe6bc98da0d6d95 Mon Sep 17 00:00:00 2001
From: Benoit Putzeys <157973952+bputzeys@users.noreply.github.com>
Date: Tue, 4 Nov 2025 18:09:38 +0100
Subject: [PATCH 2/2] Update pyproject.toml

---
 pyproject.toml | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pyproject.toml b/pyproject.toml
index 14704b54..92f757a5 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "helical"
-version = "1.4.8"
+version = "1.4.9"
 authors = [
   { name="Helical Team", email="support@helical-ai.com" },
 ]