Estimating Transcript Abundance
The BAM isn’t the final file
- BAM files give the loca)on of mapped reads;
- But, per individual, how many reads should be considered as from any par)cular gene?
- The count table represents this;
A common and logical method to estimate transcript abundance across a reference transcriptome using RNA-seq data is to count the number of reads that map uniquely to each transcript. Reads that map to multiple contigs or transcripts may provide ambiguous information and therefore introduce more noise than is desirable in some contexts. If, on the other hand, alternative splice variants are separate features in the reference, one would expect multiply mapped reads and may want to count them, but in an unbiased manner. In any case, it is straightforward to count the number of reads aligned to each reference feature from a SAM file, write a tabular file containing this information, and use this file for differential gene expression analysis.
Count-table Example:
