ren-lab / snp-selex Goto Github PK

Hi, thanks for the great tool and resources on TF binding! I intend to use the 94 deltaSVM models to predict the effect of several de novo SNV, and i'm puzzled about the deltasvm folder. It seems that the scripts in this folder are used to train novel model (and they need the large file you mentioned in readme), and seems that the models presented in the paper are not stored here. Is there a way to directly apply your 94 model on new SNV in a way like "python model.py file_storing_SNV_sequence"? Maybe i missed it, but i failed to find similar commands in the snakefile. Thank you very much for your help!

Hi,

Thanks for making the codes publicly available.

I am trying to re-train the svm model with your codes, and I have downloaded the 7443 fastq files from https://www.ebi.ac.uk/ena/browser/view/PRJEB9797,

but when I try to run snakemake, there is an error message saying zerocycle file is missing:

Building DAG of jobs...
MissingInputException in line 56 of /rwthfs/rz/cluster/hpcwork/rwth0776/snp-selex/deltasvm/Snakefile:
Missing input files for rule fastqTofasta:
fastq/ZeroCycle_TGTTAA40NACT_0_0.fastq.gz

so where can I download these files?

Thanks and best,
Zhijian Li