🧬 DNBelab C Series HT scVDJ 分析参数

🔬 主分析流程 (run)

📊 用法

$ dnbc4tools vdj run -h
usage: dnbc4tools vdj run [OPTIONS] 

optional arguments:
  -h, --help            show this help message and exit

Input Files:
  Choose ONE input method: either --fastqs (directory) OR individual FASTQ files (-1 and -2).

  --fastqs <DIR>        Input directory containing paired-end FASTQ files. The pipeline automatically detects Read1/Read2 files. Example: ./fastq_dir
  -1, --fastq1 <FILE> [<FILE> ...]
                        Read1 FASTQ file(s) (supports wildcards and comma-separated lists). Example: sample1_L01_R1.fastq.gz,sample1_L02_R1.fastq.gz
  -2, --fastq2 <FILE> [<FILE> ...]
                        Read2 FASTQ file(s) (supports wildcards and comma-separated lists). Must match --fastq1 order. Example: sample1_L01_R2.fastq.gz,sample1_L02_R2.fastq.gz

Basic Settings:
  -n, --name <STR>      Unique identifier for the sample (e.g., sample1). Used for naming output files and reports.
  -r, --ref <REF>         Reference database: 'human'/'mouse' (case-insensitive) or path to a custom reference directory containing reference.json. Examples: human | mouse | ./custom_vdj_ref
  -c, --chain <STR>     VDJ receptor type: 'IG' (B-cell receptors) or 'TR' (T-cell receptors).
  -o, --outdir <DIR>    Output directory for results and reports [default: current directory]. Example: ./output
  -t, --threads <INT>   Number of CPU threads for parallel processing [default: all available cores] (e.g., 16).
  -s, --beadstrans <FILE>
                        RNA analysis singlecell.csv file for filtering cells and merging beads information. When not provided, all cells will be kept by default (equivalent to --keep_all_cells).

Library Settings:
  Auto-detection is recommended for dark cycles. Available modes include "R1" and "unset".
  For multiple files, ensure consistent settings across all inputs.
  customize: Specify sequence structure patterns for parsing.

  --darkreaction <STR>  Dark cycle setting for VDJ library [default: auto]. Use 'R1' if dark cycles occur in Read1; otherwise leave as 'auto' or 'unset'.
  --customize <STR>     Sequence structure patterns, format: <type>,<read>:<start>-<end> separated by ';'. Types include: cb (cell barcode), umi (UMI) R1/R2 (sequence). Example:
                        "cb,R1:1-10;cb,R1:11-20;umi,R1:21-30;R1,R1:31-120;R2,R2:1-150"
  --enrichment_primers <FILE>
                        Custom inner enrichment primers file (one primer sequence per line). Required when using a custom reference database.

Analysis Settings:
  --keep_all_cells      Keep all cells in analysis without RNA data filtering. If --beadstrans is not provided, this behavior is enabled by default.
  --r2_only             Only use R2 reads for VDJ assembly. Manual setting required because Read1 assembly requirements cannot be auto-detected.
  --sample_read_pairs <INT>
                        Subsample the specified number of read pairs from the input FASTQ files (e.g., 1000000).

📝 参数说明

🔴 必需参数

⚠️ 成功分析必须指定的基本参数

--name sample_VDJ_001

# 使用内置的人类参考数据库
--ref human

# 使用自定义参考数据库
--ref ./custom_vdj_ref

# 分析T细胞受体
--chain TR

# 分析B细胞受体
--chain IG

🟢 输入文件参数

📁 选择一种输入方式：基于目录 OR 单独指定文件

--fastqs ./VDJ_fastq_dir

--fastq1 sample1_L01_R1.fastq.gz,sample1_L02_R1.fastq.gz

--fastq2 sample1_L01_R2.fastq.gz,sample1_L02_R2.fastq.gz

⚠️ 输入方式选择：

• 🔸 方式1： 使用--fastqs指定包含配对文件的目录。
• 🔸 方式2： 使用-1, --fastq1和-2, --fastq2分别指定R1和R2文件。

⚠️ 重要提示： 参数下所有文件必须来自同一文库，测序模式和暗反应设置保持一致，不同文库的数据不能合并分析。

🟢 基本设置参数

--outdir ./VDJ_analysis_output

--threads 16

--beadstrans ./RNA_analysis_output/outs/singlecell.csv

🟢 文库设置参数

# Read1存在暗循环
--darkreaction R1

# 标准VDJ文库配置示例
--customize "cb,R1:1-10;cb,R1:11-20;umi,R1:21-30;R1,R1:31-120;R2,R2:1-150"

GTCCTCGGTGGCCTCCACGTG
AGCACCTGGGGCCTCGGCCAC
CCTGGACTCCTGGGCCCCAG

🚩 分析设置参数

--sample_read_pairs 10000000