The platform includes the analysis piplines after sequenced data.

Background

Heterogeneity is one of the most important characteristic of tumors. However, the current researches are focused on cancer cells of a certain tissue, which have revealed a complex and diverse state of cancer cells. But these studies ignored a key factor that is the complex cell ecosystem composed of tumor-related stromal cells and immune microenvironment may itself be a heterogeneous complex system like cancer cells. So the authors used single-cell transcriptome sequencing technology to analyze the microenvironment of stromal cells in lung cancer.

Protocol

In this experiment, specimens of tumors and adjacent non-malignant tissues were taken from the lesions of five lung cancer patients, which are separated and labeled as tumor core samples, tumor intermediate samples, tumor marginal samples, and normal tissues adjacent to cancer. The samples were then sequenced using single-cell transcriptome sequencing technology of 10x Genomics. Finally, perform cell cluster analysis and survival analysis. (The platform process uses only two of the samples for testing.)

Result

The authors analyzed 52,698 cells collected and identified 52 subtypes of stromal cells, including new cell subgroups that are considered to be homogeneous, as well as their heterogeneous transcription factors. By comparing with tissue sequencing data, it was found that the marker genes of lung cancer and the related mechanism of survival, as well as its immune response, etc.

Lambrechts, Diether, et al. "Phenotype molding of stromal cells in the lung tumor microenvironment." Nature medicine 24.8 (2018): 1277-1289.

v2.0: The process is automated, the code is uploaded to the CRS platform (ID: 117), the software is uploaded to the SRS platform (ID: 63), and the reference genome is uploaded to the DRS platform, making it possible to combine with other processes.

v1.1: Add cell type annotation funtion.

v1.0: Process construction, complete the basic process of data quality control, comparison, and differential gene calculation.

Instruction for Upload File

1. A compressed file containing the data to be analyzed

(1). If the data is not stored on this platform, upload the data from Website to use the analysis protocol. Please compress all data into a single file. (supported compression formats: rar, zip, gz, tar.gz.)

(2). If the data to be analyzed has been stored on this platform (DRS or Local), please upload any content compressed file that conforms to the format.

2. Data basic information and project parameter json file.

Support to modify the parameters as follows:

(1). "dataId"

If the data has been stored in this platform (DRS), DRS id is needed.

(2). "dataFrom": "DRS", "Local", "Web"

Source of the analyzed data.

(3). "SampleGroup": "Case", "Control"

Sample classification

(4). "SampleName"

Sample name

(5). "OmicLocation": "RNA", "DNA"

Sequencing category

(6). "Pipeline"

Which pipeline to used. (fixed in this project)

(7). "SeqRegion": "RRBS", "WGBS", "Target"

The sequencing area of this sample is RRBS (MspI enzyme), WGBS, or Targeted bisulfite sequencing.

(8). "SeqPlatform": "Illumina", "SOLiD", "Roche454"

Sequencing platform

(9). "SeqReadType": "Single_end", "Paired_end"

The sample is single-end sequencing or paired-end sequencing.

(10). "FileType": "sra", "fq", "fq.gz", "fastq", "fastq.gz"

The sample file format.

(11). "fileName"

The name of the data to be analyzed (including file type).

If the data contains only a single file (such as single-end sequencing data or .sra suffix file), just fill in the file name; if the data contains multiple files (such as paired-end sequencing data file_{1,2}.fq. gz), you need to fill in "fillName" and "FilenameR2" at the same time.

(12). "FilenameR2"

If the data contains multiple files (for example: paired-end sequencing data file_{1,2}.fq.gz), you need to fill in "fillName" and "FilenameR2" at the same time.

(13). "AlignmentSoftware"

The comparison software used in the process. Default is Hisat2.

(14). "Race"

Sequencing samples. The default is the whole population.

(15). "Do_QC": "yes", "no"

Whether to perform data quality control. It is recommended to choose "yes" unless the data has been trimmed.