10xgenomics / cellranger-atac Goto Github PK

View Code? Open in Web Editor NEW

16.0 6.0 9.0 3.76 MB

License: Other

Makefile 0.26% Shell 2.31% Python 72.61% C++ 9.25% HTML 1.15% Go 11.73% C 2.70%

cellranger-atac's Introduction

Building Cell Ranger ATAC 1.2.0

Build dependencies

This package has been built internally using:

Python 2.7.13
clang 7.0 (gcc/g++ should also work)
go 1.12

Setting up the environment and build dependencies

Environment dependencies can be found in the Ranger ATAC 1.2.0 package (https://support.10xgenomics.com/developers/software/downloads/latest)
- The Ranger package includes a build of the Martian platform (v3.2.4), which is open source. For more information, go to (http://martian-lang.org).
- The Ranger package includes a build of miniconda, which contains required python executables and packages.

Example setup of build dependencies on Ubuntu 20.04

# use gcc/g++ in this case to support GLIBCXX_3.4.25 in pre-built ranger package
sudo apt-get install make gcc g++ pkg-config

# Add golang to path
wget https://golang.org/dl/go1.12.17.linux-amd64.tar.gz
tar xf go1.12.17.linux-amd64.tar.gz
export PATH=/path/to/go/bin:$PATH
export GOROOT=/path/to/go

# Source the Ranger package for python, the Martian Environment and binary dependencies
source /path/to/ranger-1.2.0-ATAC/sourceme.bash

Build command

make

Running Cell Ranger ATAC

Setting up the Runtime dependencies for Cell Ranger ATAC

source /path/to/open-source/cellranger-atac/sourceme.bash

Note about Loupe

The binaries required to generate Loupe Cell Browser (.cloupe) are not included in this repository or in the binary dependencies package Ranger. The necessary binaries, including crconverter, bamtofastq, etc. can be obtained from an existing binary version of Cell Ranger ATAC by running:

cp /path/to/cellranger-atac/cellranger-atac-cs/1.2.0/lib/bin/crconverter /path/to/open-source/cellranger-atac/lib/bin/

Support

We do not provide support for building and running this code.

The officially supported release binaries are available at: (https://support.10xgenomics.com/single-cell-atac/software/downloads/latest)

cellranger-atac's People

Contributors

Stargazers

Watchers

Forkers

ragnardanneskjold harvardinformatics mengxiao procha2 smenon2 huynhnpt jsgro sarangian sukses24

cellranger-atac's Issues

About raw_barcode_count in mark_duplicates

Thank you for the development of cellranger-atac pipeline. My colleague @haowenz and I are trying to understand the output from cellranger-atac, especially for the deduplication. During the exploration, we found that some of the most common barcodes in the dup groups did not agree with the frequency in the sample, thus the wrong barcodes became the representatives for the dup groups. After examining the code, we think the barcode_whitelist should be sorted to match the count:

https://github.com/10XGenomics/cellranger-atac/blob/main/mro/atac/stages/processing/mark_duplicates/__init__.py#L415-L421

After adding something like "barcode_whitelist = sorted(list(barcode_whitelist))" we got expected results. We are wondering if this is a valid fix. Is this a bug in CellRanger-atac? Thank you.

Cellranger-atac error for incorrect barcode

Need a bam file without remove duplicate

I am working on the dataset of Single Cell Multiome ATAC + Gene Exp, and i expect a bam file of scATAC contain duplicate or UMI, how I can get that?

Title: CellRanger-ARC for Arbidosis failed with error

I am running CellRanger-ARC for Arabidopsis and I am getting the following error:

2023-06-02 14:40:02 [runtime] (failed)          ID.D3.SC_ATAC_GEX_COUNTER_CS.SC_ATAC_GEX_COUNTER._SC_ATAC_GEX_ANALYZER._PEAK_ANNOTATOR.ANNOTATE_PEAKS

[error] Pipestance failed. Error log at:
D3/SC_ATAC_GEX_COUNTER_CS/SC_ATAC_GEX_COUNTER/_SC_ATAC_GEX_ANALYZER/_PEAK_ANNOTATOR/ANNOTATE_PEAKS/fork0/join-uc5dd7a4574/_errors

Log message:
Traceback (most recent call last):
File "/data/rc/apps/rc/software/CellRanger-ARC/2.0.0/external/martian/adapters/python/martian_shell.py", line 659, in _main
stage.main()
File "/data/rc/apps/rc/software/CellRanger-ARC/2.0.0/external/martian/adapters/python/martian_shell.py", line 629, in main
lambda: self._module.join(args, outs, chunk_defs, chunk_outs)
File "/data/rc/apps/rc/software/CellRanger-ARC/2.0.0/external/martian/adapters/python/martian_shell.py", line 589, in _run
cmd()
File "/data/rc/apps/rc/software/CellRanger-ARC/2.0.0/external/martian/adapters/python/martian_shell.py", line 629, in <lambda>
lambda: self._module.join(args, outs, chunk_defs, chunk_outs)
File "/data/rc/apps/rc/software/CellRanger-ARC/2.0.0/mro/atac/stages/analysis/annotate_peaks/init.py", line 105, in join
gene_id_name_map = build_gene_id_name_map(ref_mgr)
File "/data/rc/apps/rc/software/CellRanger-ARC/2.0.0/mro/atac/stages/analysis/annotate_peaks/init.py", line 60, in build_gene_id_name_map
fields.attrs.get("gene_name", fields.attrs["gene_id"])
File "pybedtools/cbedtools.pyx", line 392, in pybedtools.cbedtools.Interval.attrs.get
File "pybedtools/cbedtools.pyx", line 180, in pybedtools.cbedtools.Attributes.init
ValueError: need more than 1 value to unpack

I have tried the following to resolve the issue:

I have updated CellRanger-ARC to the latest version.
I have checked the input data to make sure that it is in the correct format.
I have checked the software used to run CellRanger-ARC to make sure that it is up to date and that it has all of the necessary dependencies.
I am still unable to resolve the issue.

Please help me to troubleshoot this issue.

Thank you.

Odd fragment.tsv.gz files

Hi!

I inspected my fragment files using gzip -dc fragments.tsv.gz | head and got the following output:

I believe this is causing problems when reading the file into python's pycistopic package. Other users show that their output for the same line of code is as follows:

Any way of removing that information at the start of the fragment file?

Thanks,
Susana

cellranger-atac argg balcklist ratio Error

As this issue suggested, stuart-lab/signac#889:
cellRanger released the new version(2.0.0) to preprocess ATAC-seq data. and it announced that the annotation track 'blacklist.bed' has been removed, so the column "blacklist_region_fragments" in the singlecell.csv output for the corresponding tracks are all zero.

but in the singlecell.csv , the output of cellranger-atac agrr , the blacklist ratio is calculated.

most importantly, the blacklist ratio seems wrong for some reason. I found this problem by running the same sample with cellranger-atac V1.2 and v2.0.0, which reported totally different blacklist ratio.

preflight error UTF-8

Hello all,
I am trying to run the cellranger atac pipeline on SRA derived fastqs. I continuously run into a preflight error:

UnicodeDecodeError: 'utf-8' codec can't decode byte 0x8b in position 1: invalid start byte

Is there a fix for this? I tested multiple fasts to rule out input problems.

Non-nuclear read pairs

I'm using version 2.0.0. In the summary.csv file, there's a value reported for Non-nuclear read pairs. How is this number calculated? I can see in the html that it describes cell barcodes with >Q30 but i'm trying to reproduce this number using the singlecell.csv metrics and can't figure it out.