celescope 2.3.0.dev2__tar.gz → 2.3.1.dev0__tar.gz

{celescope-2.3.0.dev2/celescope.egg-info → celescope-2.3.1.dev0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: celescope
-Version: 2.3.0.dev2
+Version: 2.3.1.dev0
 Summary: Single Cell Analysis Pipelines
 Home-page: https://github.com/singleron-RD/CeleScope
 Author: zhouyiqi

{celescope-2.3.0.dev2 → celescope-2.3.1.dev0}/celescope/__init__.py

@@ -1,6 +1,6 @@
 import os
 
-__VERSION__ = "2.3.0dev2"
+__VERSION__ = "2.3.1dev"
 __version__ = __VERSION__
 
 ASSAY_LIST = [

{celescope-2.3.0.dev2 → celescope-2.3.1.dev0}/celescope/bulk_rna/starsolo.py

@@ -81,12 +81,6 @@ class Starsolo(tools_Starsolo):
         """take in raw matrix, only keep barcodes in the input file, and convert barcodes to sample names"""
         matrix = CountMatrix.from_matrix_dir(self.raw_matrix)
 
-        for barcode in self.barcode_sample:
-            if barcode not in matrix.get_barcodes():
-                sys.stderr.write(
-                    f"WARNING: barcode:{barcode} {self.barcode_sample[barcode]} not found in raw matrix!\n"
-                )
-                self.barcode_sample.pop(barcode)
         filtered = matrix.slice_matrix_bc(self.barcode_sample.keys())
         filtered.to_matrix_dir(self.filtered_matrix)
         samples = [self.barcode_sample[bc] for bc in filtered.get_barcodes()]
@@ -135,6 +129,8 @@ class Cells(Step):
 
         df_counts.loc[:, "mark"] = "UB"
         df_counts.loc[bcs, "mark"] = "CB"
+        df_counts.fillna(0, inplace=True)
+        df_counts = df_counts.astype({"UMI": int, "countedU": int})
         df_counts.to_csv(self.counts_file, sep="\t", index=True)
 
         self.add_metric(
@@ -176,7 +172,7 @@ class Cells(Step):
         # table
         df_cells = df_counts[df_counts["mark"] == "CB"]
         df_cells["Sample"] = df_cells.index.map(lambda x: barcode_sample[x])
-        df_cells["Genes"] = df_cells.index.map(lambda x: bc_geneNum[x])
+        df_cells["Genes"] = df_cells.index.map(lambda x: bc_geneNum.get(x, 0))
         df_cells["Barcode"] = df_cells.index
         barcode_well = {v: k for k, v in well_barcode.items()}
         df_cells["Well"] = df_cells["Barcode"].map(lambda x: barcode_well[x])

{celescope-2.3.0.dev2 → celescope-2.3.1.dev0}/celescope/tools/split_fastq.py

@@ -1,5 +1,6 @@
 """Split fastq file according to barcode. Append UMI to read name with umi_separator semicolon ':'"""
 
+import io
 from celescope.tools.step import Step, s_common
 from celescope.tools import utils
 import celescope.tools.parse_chemistry as parse_chemistry
@@ -35,8 +36,11 @@ class Split_fastq(Step):
     def split(self):
         self.fh_dict = {}
         for barcode, sample in self.barcode_sample.items():
-            self.fh_dict[barcode] = utils.generic_open(
-                f"{self.outdir}/{sample}.fq.gz", "wt"
+            file_obj = utils.generic_open(
+                f"{self.outdir}/{sample}.fq.gz", "wb", compresslevel=1
+            )
+            self.fh_dict[barcode] = io.BufferedWriter(
+                file_obj, buffer_size=16 * 1024 * 1024
             )
 
         for fq1, fq2 in zip(self.fq1_list, self.fq2_list):
@@ -51,7 +55,7 @@ class Split_fastq(Step):
                     umi = e1.sequence[self.pattern_dict["U"][0]]
                     name = e1.name + UMI_SEPARATOR + umi
                     self.fh_dict[corrected_bc].write(
-                        utils.fastq_line(name, e2.sequence, e2.quality)
+                        utils.fastq_line(name, e2.sequence, e2.quality).encode()
                     )
 
     def run(self):

{celescope-2.3.0.dev2 → celescope-2.3.1.dev0/celescope.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: celescope
-Version: 2.3.0.dev2
+Version: 2.3.1.dev0
 Summary: Single Cell Analysis Pipelines
 Home-page: https://github.com/singleron-RD/CeleScope
 Author: zhouyiqi