conda installation requirements.txt

SILVER-seq on 128 serum samples from healthy and cancer (recurrent and non-recurrent) patients.

Linked publication: Zixu Zhou, et al., Extracellular RNA in a single droplet of human serum reflects physiologic and disease states.  PNAS, 2019, 116:19200–19208.

Linked accessions: GEO - GSE131512 SRA - SRP198979 BioProject - PRJNA543872

1. Get accession list (SRR_Acc_List.txt) and run table (SraRunTable.txt) from https://trace.ncbi.nlm.nih.gov/Traces/study/?acc=SRP198979&o=acc_s%3Aa
2. Pull out non-index fastq accessions SraRunTable using process_metadata.ipynb (SRR_Acc_List_NonIndex.txt)
3. Downloaded fastqs using download_data.sh

• requires fastq-dump from sra-toolkit (conda install -c bioconda sra-tools)

1. Download reference genome from Illumina iGenomes:

• wget ftp://igenome:[email protected]/Homo_sapiens/NCBI/GRCh38/Homo_sapiens_NCBI_GRCh38.tar.gz

2. Run bwa.sh to align to GRCh38 and pull out unmapped reads

• Use [bwa mem] to align (conda install -c bioconda bwa)
• Use [samtools] to pull out unmapped reads (conda install -c samtools)

1. Install kraken2 (conda install -c bioconda kraken2)
2. Download MiniKraken2_v1_8GB database (no build required)

• wget ftp://ftp.ccb.jhu.edu/pub/data/kraken2_dbs/old/minikraken2_v1_8GB_201904.tgz

3. Run kraken2.sh to classify unmapped sequences

1. Create qiime environment (https://docs.qiime2.org/2020.2/install/native/#install-miniconda)
2. Step through qiime_analysis.ipynb to move data into qiime

• Generate sample_metadata.tsv using process_metadata.ipynb

https://docs.qiime2.org/2019.10/plugins/available/sample-classifier/ https://docs.qiime2.org/2020.2/tutorials/sample-classifier/ Run qiime_analysis.ipynb to perform sample classification using Qiime plugin