Tweet
good "benchmark" or minimum bayes factor for differential expression?
I'm filtering a file with genome wide analysis of bayes factors between two stages in neuron development. What's a good minimum BF for a gene to be counted as differentially expressed? In a Nature Communications paper it seems that they used BF>20, but I'm not sure if that's just for the specific study they were doing, or if that's a generally universal value.
-
-
In general, Bayes Factors are a bit like interpreting a p-value and the choice of cutoff for "significance" is arbitrary. The Jeffrey's scale is one widely used suggested scaling of significance where K's >10 are "strong" inidcators of significant differences, and K's > 10^2 are "decisive" indicators of significance and anything in between as "very strong" indication (that is, that gene expression levels for population M2 are greater than those for population M1, for example).
In my experience, Bayes factors are not at all commonly used for differential gene expression, so you may need to convince an audience of the validity of the methods used to generate that file (not to mention explaining just what a Bayes factor is as most genomic biologists will have never heard of the term).Last edited by mbblack; 08-05-2015, 04:18 AM.Michael Black, Ph.D.
ScitoVation LLC. RTP, N.C.
-
Genomics studies in neuroscience face a special challenge due to the brain’s complexity and scarcity of samples. Mapping changes in cell type and state using conventional next-generation sequencing methods remains challenging. Advances in technologies like single-cell sequencing, spatial transcriptomics, and long-read sequencing have opened the door to deeper studies of the brain and diseases like Alzheimer’s, amyotrophic lateral sclerosis (ALS), and schizophrenia.
...07-09-2026, 11:10 AM -
Cancer survival rates have significantly increased in the last few decades in the United States, reaching a combined 70% 5-year survival rate by 2021. Behind this number, there are years of research to find new therapies, drug targets, and early detection methods. But there is one core challenge that keeps slowing down these advances, and it’s about drug resistance.
There is no single reason why many patients don’t respond to treatment as expected. Cancer is...07-08-2026, 05:17 AM -
Love this - good data definitely starts from good input, and poor input can only give relatively poor data. I particularly like the mention of Nanodrop/absorbance based methods for quantification. It's such a toss up if you'll get an accurate reading or what amounts to a randomly generated number, and a lot of library/sequencing related issues can be traced back to poor quant.07-01-2026, 11:43 AM
Comment