Single-cell RNA seq

Question

[shg018]

HI

I've been trying to learn single-cell sequencing protocol from a published paper "Single-cell RNA-seq enables comprehensive tumour and immune cell profiling in primary breast cancer", Woosung Chung et al. 2017. The paper states that "To assess the expression values of array control RNA spike-ins, reference sequences and the corresponding annotations were generated by merging three control RNA spike-ins (ThermoFisher) with the human genome reference sequences (hg19) and the GENCODE 19 annotations. The RNA reads were then aligned to the reference sequences using the 2-pass mode of STAR_2.4.0b (default parameters), and relative gene expression was quantified as transcript per million (TPM) using RSEM v1.2.17 (default parameters)".

So far I have downloaded the bulk data from SRA and converted it to fastq, I am not sure how to proceed with generating the reference and corresponding annotations? Do I just use the gtf and fasta file for hg19 when making the index that I would map the fastq files to?

Answer 1

[seidel]

Yes, you would use the hg19 fat sequence and gtf, to make an alignment index. But if you want your alignment results to contain values for the control spikes, they are saying that you'll have to supplement the hg19 reference and gtf with sequences and annotation for the control spikes. If you don't care about the spikes, you can ignore that, and the reads matching spikes will simply be unmapped.