I have two bed files of genomic coordinates. (say 12000 entries in one, 1000 in the other) I know how many unique bases in the genome are covered by each file (say 12000=15% of the genome, 1000=3% of the genome. I overlap them and get a result, say 50% of the 1000 overlap the entries in the 12000. Does anyone have suggestions on how to test the significance of this given that any overlap (not complete) is counted as overlap? I've thought of adding expanding the refernence (12000) entries by 50% of the average length of the 1000 entries, but that seems a bit to crude.
-
-
You can do a Monte carlo against randomly distributed regions, or even regions randomly distributed that are still the same distance to TSS or such. Will basically always call your result significant though, because the null model is bad. -
Sorry what does your last sentence mean? My plan is to use shuffleBed with intersectbed over 1000x and then plot the distribution of intersections and see where the actual data falls on the plot. Is that not sound?Comment
-
NGS has revolutionized clinical microbiology by offering new ways to identify and surveil pathogens without prior knowledge of the target. One popular approach is 16S rRNA and ITS amplicon sequencing, which targets specific microbial markers. But studies are increasingly incorporating shotgun metagenomics, a technique that sequences all DNA from a sample without targeting specific microbial markers. This approach allows more detailed taxonomic classification, detection...Today, 06:46 AM -
Tumors contain complex mixtures of cell populations that evolve in response to treatment and host-specific factors. Tissue biopsies provide only static snapshots of a tumor’s state. But what if we could follow that evolution over time using a minimally invasive technique?
Liquid biopsies can detect tumor-derived material circulating in the bloodstream, providing a way to study treatment response, detect emerging resistance, and discover new biomarkers. Circulating...10-14-2026, 09:55 AM
Comment