timydaley/crisphiermix

crisphiermix's Issues

Possibility of running without replicates?

Hi, I would like to apply this tool but am a bit confused about the input requirements.
My lab has performed a CRISPRi dropout screen in which we have sequenced three replicates 21 days post-infection. For the "day 0" input we are planning to use the plasmid library which we sequenced prior to infection. However, we only sequenced this once (as additional sequencing should be virtually the same). My understanding is that the LF2C from Deseq2 is supposed to be used as input to CRISPhieRmix but this would require "replicates" of the plasmid library in our case.

Am I understanding this correctly? Do you have any suggestions for how I might proceed?
Thank you!

Weird results

When I run CRISPhieRmix with "negCtrl", it returns all FDR equal to 1! Do you have any idea what is happening?!

Also, the gene names as input and output are not in the same order. I'm worried if the lists are shuffled and the results would be messed up.

@timydaley, plz help 🙏

I am getting an error that halts execution of the script.

Error is
"Error in if (abs(loglike - prevloglike)/n_obs < tol | iter > max_iter) { :
missing value where TRUE/FALSE needed"

The input is CRISPhieRmix_Simulation_l2fc.txt
control is
Simulation_l2fc_control.txt

CRISPhieRmix_Simulation_l2fc.txt