Calculated LAI is too large

[yaoxkkkkk]

Thank you for your effort on these tools, I am using EDTA (v2.2.1) to carry out de novo TE annotation. After I used EDTA output file fasta.mod.EDTA.raw/LTR/.fasta.mod.pass.list and .fasta.mod.EDTA.anno/.fasta.mod.out as input to calculate LAI, but the LAI seemed abnormal:

Chr	From	To	Intact	Total	raw_LAI	LAI
whole_genome	1	521293298	0.0515	0.0667	77.13	74.94
Chr01_hap1	1	3000000	0.0355	0.0234	100.1	97.91
Chr01_hap1	300001	3300000	0.0423	0.0256	100.1	97.91
Chr01_hap1	600001	3600000	0.0475	0.0277	100.1	97.91
Chr01_hap1	900001	3900000	0.0475	0.0285	100.1	97.91
Chr01_hap1	1200001	4200000	0.0415	0.0262	100.1	97.91

For another haplotype:

Chr	From	To	Intact	Total	raw_LAI	LAI
whole_genome	1	503369137	0.0497	0.0594	83.75	82.73
Chr01_hap2	1	3000000	0.0285	0.0232	100.1	99.08
Chr01_hap2	300001	3300000	0.0263	0.0277	94.79	93.77
Chr01_hap2	600001	3600000	0.0311	0.0299	100.1	99.08
Chr01_hap2	900001	3900000	0.0287	0.0286	100.1	99.08
Chr01_hap2	1200001	4200000	0.0287	0.0277	100.1	99.08
Chr01_hap2	1500001	4500000	0.0303	0.0293	100.1	99.08
Chr01_hap2	1800001	4800000	0.0289	0.0283	100.1	99.08
Chr01_hap2	2100001	5100000	0.0297	0.0289	100.1	99.08
Chr01_hap2	2400001	5400000	0.0197	0.0284	69.32	68.30
Chr01_hap2	2700001	5700000	0.0214	0.0315	67.83	66.81
Chr01_hap2	3000001	6000000	0.0212	0.0304	69.72	68.70

Here is the log of LAI:

######################################
### LTR Assembly Index (LAI) beta3.2 ###
######################################

Developer: Shujun Ou

Please cite:

Ou S., Chen J. and Jiang N. (2018). Assessing genome assembly quality using the LTR Assembly Index (LAI). Nucleic Acids Res. gky730: https://doi.org/10.1093/nar/gky730

Parameters: -t 64 -genome /dssg/home/acct-jiang.lu/jiang.lu/.yaoxk_dir/genome_denovo/SH/04-asm/03-final/hap2/Vitis_amurensis_ShuangHong_hap2.fasta -intact /dssg/home/acct-jiang.lu/jiang.lu/.yaoxk_dir/genome_denovo/SH/05-ant/01-repeat/hap2/repeat_library/TE/Vitis_amurensis_ShuangHong_hap2.fasta.mod.EDTA.raw/LTR/Vitis_amurensis_ShuangHong_hap2.fasta.mod.pass.list -all /dssg/home/acct-jiang.lu/jiang.lu/.yaoxk_dir/genome_denovo/SH/05-ant/01-repeat/hap2/repeat_library/TE/Vitis_amurensis_ShuangHong_hap2.fasta.mod.EDTA.anno/Vitis_amurensis_ShuangHong_hap2.fasta.mod.out


Tue May 21 13:29:55 CST 2024	Dependency checking: Passed!
Tue May 21 13:29:55 CST 2024	Calculation of LAI will be based on the whole genome.
				Please use the -mono parameter if your genome is a recent ployploid, otherwise high identity between LTR homeologues will overcorrect raw LAI scores and result in low LAI.
Tue May 21 13:29:55 CST 2024	Estimate the identity of LTR sequences in the genome: standard mode
Tue May 21 13:35:26 CST 2024	The identity of LTR sequences: 94.3636850089836%
Tue May 21 13:35:26 CST 2024	Calculate LAI:

						Done!

Tue May 21 13:35:35 CST 2024	Result file: Vitis_amurensis_ShuangHong_hap2.fasta.mod.out.LAI

				You may use either raw_LAI or LAI for intraspecific comparison
				but please use ONLY LAI for interspecific comparison

Looking forward to you advice!

[yaoxkkkkk]

I have changed the input file as .fasta.mod.EDTA.anno/.mod.EDTA.TEanno.out, here is the result file:

Chr	From	To	Intact	Total	raw_LAI	LAI
whole_genome	1	521293298	0.0515	0.0605	85.11	83.61
Chr01_hap1	1	3000000	0.0355	0.0214	100.1	98.60
Chr01_hap1	300001	3300000	0.0423	0.0235	100.1	98.60
Chr01_hap1	600001	3600000	0.0475	0.0256	100.1	98.60
Chr01_hap1	900001	3900000	0.0475	0.0267	100.1	98.60
Chr01_hap1	1200001	4200000	0.0415	0.0247	100.1	98.60
Chr01_hap1	1500001	4500000	0.0421	0.0275	100.1	98.60
Chr01_hap1	1800001	4800000	0.0422	0.0273	100.1	98.60
Chr01_hap1	2100001	5100000	0.0435	0.0293	100.1	98.60
Chr01_hap1	2400001	5400000	0.0296	0.0241	100.1	98.60
Chr01_hap1	2700001	5700000	0.0242	0.0293	82.67	81.17
Chr01_hap1	3000001	6000000	0.0294	0.0312	94.45	92.95
Chr01_hap1	3300001	6300000	0.0210	0.0304	68.95	67.45
Chr01_hap1	3600001	6600000	0.0232	0.0315	73.63	72.13
Chr01_hap1	3900001	6900000	0.0271	0.0328	82.54	81.04
Chr01_hap1	4200001	7200000	0.0275	0.0361	76.26	74.76

[yaoxkkkkk]

I realised that EDTA seems to mask only long TEs (>=1 kb), maybe I should use RepeatMasker additionally to abtain the .out file?

[yaoxkkkkk]

OK I have tried, the results are similar... I don't know where goes wrong. I noticed that the Intact column is the same, but the Total column is slightly different.

Chr	From	To	Intact	Total	raw_LAI	LAI
whole_genome	1	521293298	0.0515	0.0682	75.46	73.50
Chr01_hap1	1	3000000	0.0355	0.0261	100.1	98.14
Chr01_hap1	300001	3300000	0.0423	0.0278	100.1	98.14
Chr01_hap1	600001	3600000	0.0475	0.0294	100.1	98.14
Chr01_hap1	900001	3900000	0.0475	0.0291	100.1	98.14
Chr01_hap1	1200001	4200000	0.0415	0.0271	100.1	98.14
Chr01_hap1	1500001	4500000	0.0421	0.0304	100.1	98.14
Chr01_hap1	1800001	4800000	0.0422	0.0303	100.1	98.14
Chr01_hap1	2100001	5100000	0.0435	0.0331	100.1	98.14
Chr01_hap1	2400001	5400000	0.0296	0.0290	100.1	98.14
Chr01_hap1	2700001	5700000	0.0242	0.0347	69.93	67.97
Chr01_hap1	3000001	6000000	0.0294	0.0364	80.82	78.86
Chr01_hap1	3300001	6300000	0.0210	0.0359	58.34	56.38
Chr01_hap1	3600001	6600000	0.0232	0.0377	61.62	59.66
Chr01_hap1	3900001	6900000	0.0271	0.0401	67.46	65.50
Chr01_hap1	4200001	7200000	0.0275	0.0432	63.75	61.79
Chr01_hap1	4500001	7500000	0.0253	0.0431	58.69	56.73

[oushujun]

Hello,

Sorry for the delayed response. Your total LTR estimation seems off. It's unlikely the total LTR content in your genome is 6.82% when you have 5.15% of the genome is intact LTR. Please double check.

Thanks!
Shujun

[yaoxkkkkk]

Your total LTR estimation seems off

Should I rerun the whole pipeline?

[oushujun]

That’s the easiest solution. Basically you need to figure out why total LTR is that low and make sure this is not due to error. Shujun

…

On Tue, Aug 13, 2024 at 10:26 PM Xiukun Yao ***@***.***> wrote: Your total LTR estimation seems off Should I rerun the whole pipeline? — Reply to this email directly, view it on GitHub <#171 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABNX4NGLCLYDHVJHKL6WQQDZRK54JAVCNFSM6AAAAABIA64T66VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEOBXG4YTIMZZGQ> . You are receiving this because you commented.Message ID: ***@***.***>