Trouble with scan1snps in haploid cross

[Gabe Rangel]

Good afternoon,

I have been using qtl2 for malaria parasite cross analysis. I have successfully completed all the tutorial scripts, including using the scan1snps command, but I cannot successfully run the scan1snps command with my own data. I continue to get the following error: 

>Error in .alleleprob_to_snpprob(genoprobs[[uchr]], sdp, interval, on_map) :

      Not compatible with requested type: [type=NULL; target=integer].

****

Below is the script I am using: 

>library(qtl2)
>HB3xDd2<-read_cross2("HB3xDd2.yaml")
>map <- insert_pseudomarkers(HB3xDd2$gmap, step=0, cores=0)
>pr <- calc_genoprob(HB3xDd2, map, error_prob=0.002, cores=0)
>out <- scan1(pr, HB3xDd2$pheno, cores=0)
>query_variants<-create_variant_query_func(dbfile = "./HB3xDd2_variant.db")
>variants <- query_variants(3, .7, 1)
>out_snps<-scan1snps(
  pr,
  HB3xDd2$pmap,
  HB3xDd2$pheno,
  query_func = query_variants,
  chr = 3,
  start=.7,
  end=1,
  cores = 0,
  keep_all_snps = T
)

The read-in files are here: https://drive.google.com/drive/folders/1mvdaB4vL2omjos_kdrb1T5fRbjjhQYJN?usp=sharing

Any and all help would be greatly appreciated! 

Cheers,

Gabe Rangel

[Karl Broman]

I think the problem is that the SNP database, or the query function, needs to be revised to include SDP information: the "strain distribution pattern" (as an integer) for the founder strains.

But this doesn't really apply in your cross, as I understand it, where you have haploids with two alleles?

I don't think scan1snps is relevant for this sort of cross; it's potentially useful for collapsing multi-parent populations to binary SNPs, but it won't add any information for your cross, over what is in the observed genotype data and the results of scan1().

karl