Contains supplementary scripts and files used to benchmark and test sTELLeR (https://github.com/kristinebilgrav/sTELLeR)
The files include two fasta files; one containing only L1s (L1HS, L1preTa, L1PA2, L1PA6, L1) and Alus (AluJb, AluY, AluJr, AluSp, AluSq, AluJo, AluYb8, AluYa5, AluSx) (TEfastasequences.fa) and another (TEsequences_SVA_HERV_ALU_L1.fasta) also containing SVA and HERV sequences. Either can be used in running sTELLeR.
Additionally there is a small simulated bam file containing TE insertions and its reference file (masked chr22 with removed masked parts - zip file. Need to use this for correct reference base. Unzip with tar -xvzf to use) which can be used for testing.
A txt file containing the correct TE positions (testbamTRUTH.txt) can be used for verification of correct installment and run.
The TruthSet folder and containing files are scripts used to generate the truthset used in benchmarking.
Example run of https://github.com/kristinebilgrav/sTELLeR using the above test data:
Download all files in repository and run:
python steller/steller.py --ref HG38_chr22.fa --TE_fasta fasta/TEfastasequences.fa --bam testdata.bam --sr 4 --style pb -o testdata_res -mr 80
Results will be given in testdata_res_repeats.vcf and correct results can be verified by comparing to the provided file testbamTRUTH.txt - a total of 7 variants should be identified.