search

zhangrengang / SubPhaser

Phase, partition and visualize subgenomes of a neoallopolyploid or hybrid based on the subgenome-specific repetitive kmers.
https://doi.org/10.1111/nph.18173
GNU General Public License v3.0
53 stars 12 forks source link

ValueError: n_components=3 must be between 0 and min #6

Open YuChrming opened 2 years ago

YuChrming commented 2 years ago

Dear Writer ! Thanks for your useful pipline, but I meet an error when use SubPhaser. My command subphaser -i groups_genome.fasta -c groups_sg.config My configue file image But I have the error image What's wrong with me ?

zhangrengang commented 2 years ago

@YuChrming, Your genome has too few differential kmers between subgenomes. It may be artifactual or natural. My suggestions (refer to the Supplementary Material):

  1. Check the HiC contact map whether there are too many switch errors between subgenomes. If not, it is natural.
  2. If natural, it may be because 1) there are too many recombinations between subgenomes after hybridization, or 2) their progenitors are too close (possible autopolyploid) to distinguish. In these cases, subgenome phasing is meaningless and subphaser will not work.
  3. Another possibility is that your genome is hybridizated from an autotetraploid and a distinct diploid, and you can try -baseline -1 to compare the most and the least abundance of kmers.
YuChrming commented 2 years ago

Thanks for your quick reply! My Hi-C heatmap is not bad, at least in my opinion. image Then, I try -baseline -1 but the error is still. image May it is highly homologous? But it should be an allotriploid evidenced by many literatures.

zhangrengang commented 2 years ago
  1. Have you removed the check point files (tmp/*ok) before re-run or used the -overwrite option?
  2. What are the evidences for allo?