LAI is much lower than raw_LAI

[bitezhu]

Hi，shujun
I am running the LAI index on a newly assembled genome (less than 200Mb). The contiguity is pretty good with N50>1.2Mb.
I have the following results :
Chr From To Intact Total raw_LAI LAI
whole_genome 1 178823252 0.0298 0.1554 19.16 13.53

I put my full commands here:
~/toolkit/LTR_Finder/source/ltr_finder -D 15000 -d 1000 -L 7000 -l 100 -p 20 -C -M 0.9 genome_10k_pilon.fa > TAIR10.finder.scn
~/toolkit/gt-1.5.10-Linux_x86_64-64bit-complete/bin/gt suffixerator -db genome_10k_pilon.fa -indexname 248_gt -tis -suf -lcp -des -ssp -sds -dna
~/toolkit/gt-1.5.10-Linux_x86_64-64bit-complete/bin//gt ltrharvest -index 248_gt -minlenltr 100 -maxlenltr 6000 -mindistltr 1500 -maxdistltr 25000 -similar 95 -mintsd 4 -maxtsd 6 -motif tgca -vic 10 -seqids yes > genome.fa.harvest.scn
~/toolkit/LTR_retriever-master/LTR_retriever -genome genome_10k_pilon.fa -inharvest genome.fa.harvest.scn -infinder TAIR10.finder.scn -threads 24

The difference between raw_LAI and LAI really confuses me. According to the equation in your paper, 'LAI = raw LAI + 2.8138 × (94 – whole genome LTR identity). ', it seems raw_LAI should always be smaller than LAI. So I wonder what causes the drop of LAI value.

Thanks in advance for your reply!

[oushujun]

@bitezhu It depends on how active were the LTR retrotransposons in your genome. The LTR identity can approximate this value. In the equation you referenced, if LTR identity goes beyond 94, the corrected LAI will be lower than the raw LAI. This value is recorded in the *.iden file. Overall LAI of 13.5 is not bad at all.

Best,
Shujun

[bitezhu]

@oushujun So higher LTR identity means this species went through LTR expansion recently?

[oushujun]

@oushujun So higher LTR identity means this species went through LTR expansion recently?

Yup.

[xingjianfeng100]

@bitezhu It depends on how active were the LTR retrotransposons in your genome. The LTR identity can approximate this value. In the equation you referenced, if LTR identity goes beyond 94, the corrected LAI will be lower than the raw LAI. This value is recorded in the *.iden file. Overall LAI of 13.5 is not bad at all.

Best, Shujun

#83 (comment)

@oushujun Then may I ask, which LAI value might I use for interspecific comparison? Should I choose the raw LAI for interspecific comparison in this case?

[oushujun]

@xingjianfeng100 please read the LAI manual or the paper :)

[xingjianfeng100]

@oushujun Thank you, well, in normal situations, the LAI is robust for interspecific comparison; actually I would like to know, when LTR identity goes beyond 94, does it mean adjusted LAI is a overcorrection index, and we should abandon it in this situation?