wassermanlab / jaspar-ucsc-tracks Goto Github PK

Hi,
Thank you for the development of JASPAR-UCSC-tracks

I am very interested in using this program. However, I try to install the program by the bash execution of the script install-pwmscan.sh. However, the following error shows every time:

gcc -fPIC -O3 -std=gnu99 -W -Wall -o hashtable.o -c hashtable.c
make: gcc: Command not found
make: *** [Makefile:42: hashtable.o] Error 127

Could you help me?

Thank you!

It would be great if you could also provide tfbs prediction tracks for mm10

While comparing

http://expdata.cmmt.ubc.ca/JASPAR/downloads/UCSC_tracks/2020/JASPAR2020_danRer11.bb
http://expdata.cmmt.ubc.ca/JASPAR/downloads/UCSC_tracks/2022/JASPAR2022_danRer11.bb

I find that the 2022 bigbed files display the matrix identifiers while those for 2020 display the TF name, at least as displayed in IGV, as pictured below.

I expect this is in error.

@oriolfornes

Shouldn't this part from the fetch* script not be "from jaspar2pfm.py" rather than from "jaspar2meme.py"?

parser.add_option("-p", action="store", type="string", dest="profiles_dir", help="Profiles directory (from jaspar2meme.py)", metavar="<profiles_dir>")

In JASPAR UCSC tracks I read that scores in the bigbed files are p-values which have benn

(scaled between 0-1000, where 0 corresponds to p-value = 1 and 1000 to p-value ≤ 10-10)

Can I use R's rescale function to recover the p-values from the scores? For instance, a score of 950 comes from a p-value of .05

library(scales)
rescale(950,c(1,10**-10),c(0,1000))
.05

In any case, I am having trouble effectively interpreting the p_value.

In PWMScan: a fast tool for scanning entire genomes with a position-specific weight matrix I read

The P-value of a PWM score x is defined as the probability that a random k-mer sequence of the length of the PWM has a binding score ≥ x given the base composition of the genome.

I would hope to find some measure of how well the sequence at a candidate (or putative) binding site identified by PWMScan matches the motif PWM. This does not seem to provide that. Or am I mistaken?

I would like possibility of being more stringent in selection of candidates from this trace by setting a threshold on the score. However, I am hesitant to adopt this approach as thresholding on the scaled P_value could introduce a bias toward a subset of the universe of motifs. Is my reasoning suspect here?

edit: Perhaps another way of getting at this is to ask: do the motifs have the same distribution of P_values as each other?. If they do, then thresholding across the board at any given P_value should remove an equal fraction of each motif's hits. Do you know if they do?

Hello! thank you for always answering the Issues.

I have had some problems with the installation of JASPAR, but I was thinking that I finally covered it. However, when I was trying to run the example command:
./scan-sequence.py genomes/sacCer3/sacCer3.fa profiles/ -o tracks/sacCer3/ --threads 4 --taxon fungi

The error ValueError: cannot convert float NaN to an integer.

My process of installation was:
Git clone https://github.com/wassermanlab/JASPAR-UCSC-tracks
bash install-pwmscan.sh
conda env create -f ./conda/environment.yml
mv pwmscan/* JASPAR-UCSC-tracks/ (I did this due to there was an error with where is matrix_scan

run ./scan-sequence.py genomes/sacCer3/sacCer3.fa profiles/ -o tracks/sacCer3/ --threads 4 --taxon fungi

I'm using JASPAR version 1.0

Thank you so much!

The provided binaries of matrix_scan and matrix_probe do not work on any of my Linux systems. (RH7 and Ubuntu 16.0x-20.0x

Do you have a working Linux binary available?

Hi, thank you for the great resources.

I'm looking at hg38 http://expdata.cmmt.ubc.ca/JASPAR/downloads/UCSC_tracks/2022/hg38/
and found some motifs that seem unrelated to Vertebrata, for example, MA2020 is from Arabidopsis thaliana. MA1879 is from Ciona intestinalis (these are just examples and there are more).

I see the option --taxon vertebrate for hg38 in your code https://github.com/wassermanlab/JASPAR-UCSC-tracks/blob/master/scan-sequences.sh
, so I assumed that only motifs linked to vertebrates are included.

Since other genome versions like hg19 or mm10 also include the non-vertebrate motif annotation, I wondered if it's intended or if there are some mistakes.

Thank you!
Nana