HISTORY.md

-## 1.0.4 (in progress)
+## 1.0.6
+- Fix for the python3 fix.
+
+## 1.0.5
+- Fix for cb_filter with python3.
+
+## 1.0.4
+- Enable cb_histogram to be used on samples without UMIs.
+- Enable filtering of cells during `demultiplex_cells`.
+- Fix incorrect pandas.read_csv call with header=-1.
 
 ## 1.0.3 
 - Python 3 support

debian/changelog

+umis (1.0.6-1) unstable; urgency=medium
+
+  * Team upload.
+  * New upstream version
+  * debhelper-compat 12 (routine-update)
+  * Standards-Version: 4.4.1 (routine-update)
+  * Set upstream metadata fields: Bug-Database, Bug-Submit, Repository,
+    Repository-Browse.
+
+ -- Steffen Moeller <moeller@debian.org>  Sat, 18 Jan 2020 22:48:47 +0100
+
 umis (1.0.3-2) unstable; urgency=medium
 
   * Build-Depends-Arch: python3-pysam

debian/compat

-12

debian/control

@@ -3,13 +3,13 @@ Maintainer: Debian Med Packaging Team <debian-med-packaging@lists.alioth.debian.
 Uploaders: Andreas Tille <tille@debian.org>
 Section: science
 Priority: optional
-Build-Depends: debhelper (>= 12~),
+Build-Depends: debhelper-compat (= 12),
                dh-python,
                cython3,
                python3-dev,
                python3-setuptools
 Build-Depends-Arch: python3-pysam
-Standards-Version: 4.3.0
+Standards-Version: 4.4.1
 Vcs-Browser: https://salsa.debian.org/med-team/umis
 Vcs-Git: https://salsa.debian.org/med-team/umis.git
 Homepage: https://github.com/vals/umis
@@ -41,7 +41,7 @@ Description: tools for processing UMI RNA-tag data
 Package: umis-examples
 Architecture: all
 Depends: ${shlibs:Depends},
-         ${misc:Depends},
+         ${misc:Depends}
 Recommends: umis
 Description: tools for processing UMI RNA-tag data (examples)
  Umis provides tools for estimating expression in RNA-Seq data which

debian/rules

@@ -18,3 +18,8 @@ override_dh_install:
 	mv debian/python3-$(PYBUILD_NAME)/usr debian/$(PYBUILD_NAME)
 	rmdir debian/python3-$(PYBUILD_NAME)
 	find debian -type d -name __pycache__ | xargs rm -rf
+
+override_dh_auto_clean:
+	dh_auto_clean
+	rm -rf umis.egg-info
+	rm umis/utils.c

debian/upstream/metadata

@@ -3,7 +3,7 @@ Reference:
     Valentine Svensson and Kedar Nath Natarajan and Lam-Ha Ly and Ricardo
     J Miragaia and Charlotte Labalette and Iain C Macaulay and Ana Cvejic
     and Sarah A Teichmann
-   Title: "Power analysis of single-cell RNA-sequencing experiments"
+  Title: Power analysis of single-cell RNA-sequencing experiments
   Journal: Nature methods
   Year: 2017
   Volume: 14
@@ -18,3 +18,7 @@ Registry:
   Entry: OMICS_12783
 - Name: bio.tools
   Entry: NA
+Bug-Database: https://github.com/vals/umis/issues
+Bug-Submit: https://github.com/vals/umis/issues/new
+Repository: https://github.com/vals/umis.git
+Repository-Browse: https://github.com/vals/umis

setup.py

@@ -8,7 +8,7 @@ def read(fname):
 
 setup(
         name='umis',
-        version='1.0.3',
+        version='1.0.6',
         description='Package for estimating UMI counts in Transcript Tag Counting data.',
         packages=find_packages(),
         install_requires=['click', 'pysam>=0.8.3', 'pandas', 'regex', 'scipy', 'toolz'],

umis/umis.py

@@ -24,7 +24,7 @@ import numpy as np
 import scipy.io, scipy.sparse
 import click
 
-VERSION = "1.0.3"
+VERSION = "1.0.6"
 
 logging.basicConfig(level=logging.INFO)
 logger = logging.getLogger(__name__)
@@ -35,6 +35,9 @@ BARCODEINFO = {"sample": BarcodeInfo(bamtag="XS", readprefix="SAMPLE"),
                "molecular": BarcodeInfo(bamtag="RX", readprefix="UMI")}
 
 def open_gzipsafe(f):
+    if is_python3():
+        return gzip.open(f, mode="rt") if f.endswith(".gz") else open(f)
+    else:
         return gzip.open(f) if f.endswith(".gz") else open(f)
 
 def safe_makedir(dname):
@@ -75,7 +78,7 @@ def read_fastq(filename):
     if filename == "-":
         filename_fh = sys.stdin
     elif filename.endswith('gz'):
-        if is_python3:
+        if is_python3():
             filename_fh = gzip.open(filename, mode='rt')
         else:
             filename_fh = BufferedReader(gzip.open(filename, mode='rt'))
@@ -485,7 +488,7 @@ def tagcount(sam, out, genemap, output_evidence_table, positional, minevidence,
     cb_hist = None
     filter_cb = False
     if cb_histogram:
-        cb_hist = pd.read_csv(cb_histogram, index_col=0, header=-1, squeeze=True, sep="\t")
+        cb_hist = pd.read_csv(cb_histogram, index_col=0, header=None, squeeze=True, sep="\t")
         total_num_cbs = cb_hist.shape[0]
         cb_hist = cb_hist[cb_hist > cb_cutoff]
         logger.info('Keeping {} out of {} cellular barcodes.'.format(cb_hist.shape[0], total_num_cbs))
@@ -758,7 +761,7 @@ def fasttagcount(sam, out, genemap, positional, minevidence, cb_histogram,
     cb_hist = None
     filter_cb = False
     if cb_histogram:
-        cb_hist = pd.read_csv(cb_histogram, index_col=0, header=-1, squeeze=True, sep="\t")
+        cb_hist = pd.read_csv(cb_histogram, index_col=0, header=None, squeeze=True, sep="\t")
         total_num_cbs = cb_hist.shape[0]
         cb_hist = cb_hist[cb_hist > cb_cutoff]
         logger.info('Keeping {} out of {} cellular barcodes.'.format(cb_hist.shape[0], total_num_cbs))
@@ -971,9 +974,9 @@ def cb_histogram(fastq, umi_histogram):
     for read in read_fastq(fastq):
         match = parser_re.search(read).groupdict()
         cb = match['CB']
-        umi = match['MB']
         cb_counter[cb] += 1
         if umi_histogram:
+            umi = match['MB']
             umi_counter[(cb, umi)] += 1
 
     for bc, count in cb_counter.most_common():
@@ -1054,9 +1057,9 @@ def cb_filter(fastq, bc1, bc2, bc3, cores, nedit):
     ''' Filters reads with non-matching barcodes
     Expects formatted fastq files.
     '''
-
     with open_gzipsafe(bc1) as bc1_fh:
         bc1 = set(cb.strip() for cb in bc1_fh)
+
     if bc2:
         with open_gzipsafe(bc2) as bc2_fh:
             bc2 = set(cb.strip() for cb in bc2_fh)
@@ -1312,7 +1315,10 @@ def is_python3():
 @click.option('--out_dir', default=".")
 @click.option('--readnumber', default="")
 @click.option('--prefix', default="")
-def demultiplex_cells(fastq, out_dir, readnumber, prefix=""):
+@click.option('--cb_histogram', default=None)
+@click.option('--cb_cutoff', default=0)
+def demultiplex_cells(fastq, out_dir, readnumber, prefix, cb_histogram,
+                      cb_cutoff):
     ''' Demultiplex a fastqtransformed FASTQ file into a FASTQ file for
     each cell.
     '''
@@ -1321,7 +1327,9 @@ def demultiplex_cells(fastq, out_dir, readnumber, prefix=""):
     parser_re = re.compile(re_string)
     readstring = "" if not readnumber else "_R{}".format(readnumber)
     filestring = "{prefix}{sample}{readstring}.fq"
-
+    cb_set = set()
+    if cb_histogram:
+        cb_set = get_cb_depth_set(cb_histogram, cb_cutoff)
     sample_set = set()
     batch = collections.defaultdict(list)
     parsed = 0
@@ -1330,6 +1338,8 @@ def demultiplex_cells(fastq, out_dir, readnumber, prefix=""):
         parsed += 1
         match = parser_re.search(read).groupdict()
         sample = match['CB']
+        if cb_set and sample not in cb_set:
+            continue
         sample_set.add(sample)
         batch[sample].append(read)
         # write in batches to avoid opening up file handles repeatedly