ren-lab / deltasvm Goto Github PK

Hello,
I ran deltaSVM on my sequences, but it is hard for me to interpret the result. For example, the last column (preferred_allele) is None, while some values in the deltaSVM_score column is greater than 1.

Hi,

To speed up the process, I set the shell script's all the -T parameter to 16 (because trial shows only 1,4,16 allowed), however, the calculation is much slower and cannot finish after long time. Is there a possible reason for this? Thanks

Hi @yjqiu

Does deltaSVM work with the hg38 reference? Will the models be transferable?

Thanks!

Hi
I'am using both deltaSVM and QBiC-Pred (http://qbic.genome.duke.edu/makeprediction) to predict the effect of genome variants on TF binding, but for these two method, I always get opposite result for same SNP and same TF. some examples from different TF family are listed below:

• "SNP(hg19)(chr_pos_ref_alt)" "TF" "prediction result from QBiC-Pred" "deltaSVM results"
• "chr10_126278191_C_T" "ETV5" "unbound>bound (mean ETV5 bind allele T other than allele C)" "Loss "
• "chr10_115406307_A_G" "HOXA10" "bound>unbound" "Gain"
• "chr10_35057854_C_A" "ATF3" "bound>unbound" "Gain"

I think there must some problem of that, so I am reporting this issue to both of your groups. I would be very grateful for any reply.

Hi @yjqiu
I am cofused about exact meaning of "seq_binding" column in deltaSVM output, is that just gotten from a ChIP-Seq data of certain cell type or speculated from algorithm?

Hi,
I run a demo inside and see a lot of warning message in log file.
Is this normal, or is there something wrong? Thank you

INFO 2021-02-23 11:57:28: Number of threads is set to 1
INFO 2021-02-23 11:57:28: load model /media/sf_deltaSVM/deltaSVM-master/models/ETV2_eDBD_4_KO_TTCGGG40NGAG.subsample_20000.10mer.model.txt
WARN 2021-02-23 11:57:28: 'N' at (null)(19) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(19) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(19) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(19) is not a valid nucleotide. Only ACGT are allowed
INFO 2021-02-23 11:57:28: reading... 1000/23055
WARN 2021-02-23 11:57:28: 'N' at (null)(28) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(30) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(31) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(32) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(33) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(19) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(31) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(32) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(19) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(28) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(30) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(31) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(32) is not a valid nucleotide. Only ACGT are allowed
WARN 2021-02-23 11:57:28: 'N' at (null)(33) is not a valid nucleotide. Only ACGT are allowed

.......

Hi, Renlab members,

Thanks for developing such great tools to handle TFs binding prediction.

I'm wondering whether deltaSVM could support the InDels predicotion, cause the InDels always are the may reason forTFs binding difference.

Sincerely

Hi @yjqiu

Would you be able to estimate the expected runtime for deltaSVM on over 1 million SNPs? Does deltaSVM scale to this many SNPs?

Thanks!

I run the deltaSVM online using demo 3 SNPs provided online server as example. In the results table there are 2 columes: "deltaSVM_score" and "preferred_allele".Sometimes there are SNP-TF lines shows relative high delta_SVM value but the "preferred_allele" shows "None", while there are relative a little bit lower "deltaSVM_score" but "preferred_allele" shows "Gain". How to interpret this scenario? Was the "preferred_allele" determined not only by delta_SVM score but also by the score for each alternative allele？ Does the "Gain" means allelel2 bind stronger to SNP than allele1? But sometims allele2_bind is -0.706563 but allele1_bind is 0.562300, lookes the reverse but still shows 'Gain'. Thanks.

The resources files link 'http://renlab.sdsc.edu/yunjiang/deltaSVM/' is empty, running the scripts/deltasvm_subset_multi needs resources/models.weights.txt and resources/threhsolds.pbs.tsv, and resources/threhsolds.obs.tsv is in need for later script.
Can you send those file again to the resources files link or just tell us how to make those file by ourself ?

The complete set of models are not present in the all_models folder