mourisl
commented
1 year ago - This is because the same set of reads is compatible with these alleles. Therefore, these alleles could not be distinguished from the data.
- Yes, it's possible that the DRB1*04:01 is not expressed or has relatively low expression. T1K filtered the allele if its abundance is less than 15% (--frac) of the dominant allele.
- Yes, abundance_1+abundance_2 will be the gene's abundance. For homozygous case, the abundance_1 is usually about twice as large as alleles from the heterozygous case.
- What was your running command of T1K?
lucy-tian
Hi Li,
Thank you for your prompt responses regarding my previous inquiries! Hope they did not take too much of your time.
I've been using T1K for genotyping using WGS data, as well as expression quantification using RNA-seq of the same set of individuals, and I have some related questions that I would like to double-check for your opinion.
Multiple alleles for
allele_1orallele_2columns: in the case where T1K is outputting a set of alleles instead of single allele for the allele columns, is it suggesting that the abundance estimated is the same for all the alleles in the set such that T1K is not able to differentiate the true allele?For my genotype and RNA-seq expression results, there are cases where given a heterozygous situation (let's say
allele_1isDRB1*03:01andallele_2isDRB1*04:01) from WGS genotyping, expression quantification using RNA-seq of the same individual only outputDRB1*03:01forallele_1and NO detection ofallele_2. Can I interpret this situation asDRB1*04:01is not expressed at the time of RNA-sequencing?When quantifying gene expression using RNA-seq, is the total abundance of the gene the sum of
abundance_1andabundance_2? Under homozygous cases, should I double the abundance forabundance_1?There is generally much more ambiguity in the
allele_[1/2]columns when using RNA-seq as input and trying to estimate abundance at the 4 digits level. An example would be genotypeHLA-C*03:03:01from WGS hadHLA-C*03:03,HLA-C*03:126,HLA-C*03:357,HLA-C*03:370,HLA-C*03:372,HLA-C*03:418,HLA-C*03:421N,HLA-C*03:422,HLA-C*03:427,HLA-C*03:460,HLA-C*03:471,HLA-C*03:481,HLA-C*03:495,HLA-C*03:524,HLA-C*03:528,HLA-C*03:616,HLA-C*03:622when using RNA-seq. How would you recommend I approach situations like this? Since I know the true genotype isHLA-C*03:03:01, I can just interpret the ambiguity in RNA-seq as a consequence of input data quality, and the abundance level should be attributed to expression ofHLA-C*03:03:01?Sorry for posting this many questions, but I would really appreciate your feedback which comes from a more professional perspective regarding the nature of the tool. I would also be willing to explain the situations in more detail if you think a zoom meeting is more efficient.
Again, thank you for all the patience!