Simulate long deletions in alignment

Question

[bioplanet]

Hi!

I am working with STAR aligner and I wanted to ask how it treats large deletions. For example, in the read below:

@read_ID
GACATCACCTCCCACAACGAGGACTACACCATCGTGGAACAGTACGAACGCGCCGAGGGCCGCCACTCCACCGGCGGCATGGACGAGCTGTACAAGTAATGAATTAATTAAGAATGAGGTTTGCAGGAGCGGATTGCTTCGAACCGCTAAAATTTATTGTCCTCCGCGAATTAATAGCTCCTGTTGGGTCATAACGATATCTCTGGCCTCGTACTCACAAGTGGCTTTCCCCGGAGCGTGCCGTAGTTTT
+
A3AAAFFFFFFBGGGFFFGGGGGHHHHFFHHHHGHHGGHHHHHGHGGHGGGGGGGEGGGGGGGGGGHHHHHHGGGGGGGGHGCFCDGGGGFGHHHGGFFHHDFHHHHHHGHHHHHHDGH<CGDHFGHEGHGGGGGGGGGECEFGGGGGGBFFGGGFGFFFFFFFFFFFF;FFFFFFFFFFFFFFFFFFFF?FFBFF.EEFFFFFFFFFFFFFFFFFFFEFFFFFFFFFFFFFDFDFFFFDAEFAAAABFF

if I cut away part of the read (and the respective quality string), is there some specific parameters I need to tweak in order for STAR to align the "before" and "after" the deletion parts and put maybe ...... where I cut away the sequence?

My standard way of running it is:

/STAR/STAR-2.5.3a/bin/Linux_x86_64/STAR --runThreadN 10 --genomeDir ../STAR_ALIGNMENTS/CONSTRUCT_star/ --readFilesIn sample.fastq

[WouterDeCoster]

If homozygous deletions are what you are after I would suggest creating an insertion in the reference genome and use that for the alignment.

[bioplanet]

Ok, I will try to explain it a bit better:

My amplicon-seq is basically a construct of ~450nts which is sequenced using paired-end sequencing (each read has 250nts). This construct can be (potentially) modified by Cas9 in a CrispR-Cas experiment, say Cas9 could cut 5 or 10 nts somewhere but my read of course will still be 250 each, it will be only missing a part. So my question was, if, say, Cas9 cuts away from 80-86, would STAR align 1-80 and 87-end correctly and add ..... between 80 and 87? Is there some option I need to activate in order for it to happen?

Thank you!