best contig too short to make mutation

[Eric4Wu]

Hi guys, I have some issue here.

I successfully install BAMSurgeon by conda and install other Prerequisites by conda or pip

test_sv.sh successfully generate the SV_simulation.bam to my desired folder.

but no matter how i change the only one SV (only DUP) range in my sv_list.txt, it just keep popping out:

best contig too short to make mutation!

and then no succesful mutations message .

i did change the SV site from 1000bp ~ 100000bp and also the location of SV but nothing has changed.

my sequencing platform is Agilent Sureselect WES (based on illumina short read + the target site they choosed)

Is it able to process WES bam file?

[Eric4Wu]

I use debug command: --debug

I can see these mutation reads did generate but just can't put into simulation.bam

I guess this is due to WES bam ? it only contain exon regions, these reads can not meet the region I

propose, so these reads are concluded as "too short".

However, addindel.py only has insertion & deletion, insertion has to insert the seq, but is there any

option just like DUP? the only purpose I want just increase the reads/coverage of mutation target region

and then i can exam the WES CNV caller.

[adamewing]

I think you've identified the issue correctly. WES means contigs will usually be shorter than addsv needs because the capture regions aren't big. You could maybe try decreasing the parameter -l/--maxlibsize to something like 100 to allow smaller contigs but I haven't tested this with WES CNV callers.

[Eric4Wu]

Hi adamewing,

thanks for reply, I will keep trying the simulation.

so in general, BAMSurgeon is create for WGS simulation based on empirical tumor model?

would your group trying on target sequencing simulation: like NEAT & TargetSim

just curious about how was the trend in simulation feild.

Many thanks!

[Eric4Wu]

Hi adamewing,

I have one question, how do I force simulator keep contig even it is too short?

I have already use "-l 1" to specify that the max library size be 1bp. So, its criteria should be every contig

bigger than 3bp should be view as mutation and be simulated into simulated bam.

However, i still see wgsim pop out this message:

[wgsim core]: the contig length is shorted than 510, will be ignored.

How does this happen even in the maxlibsize is 1bp?

my limitation is i have to simulated the read in exon target region, these region often short and not

concatenative, if I use whole gene loci to simulate, I found only some exon will be involved in the

mutation simulation process, take TP53 as an example:

TP53 (hg19 assembly , Agilent SureSelect XT2 human all exon V6 platform) , chr17:7572906-7580759

You can notice that only around 7,577,000 bp is deleted, other distribution still not change.

What I want is that every target exon region is DEL 0.2

Which parameter should I use to force wgsim core take every contig into simulation?

Many Thanks!!!

[Eric4Wu]

Hi, adamewing,

the wgsim_core message is like this :

[wgsim] seed = 1513056176 [wgsim_core] calculating the total length of the reference sequence... [wgsim_core] 1 sequences, total length: 170 [wgsim_core] skip sequence 'target' as it is shorter than 510!

I hope no sequence will be skip!!

---

oww, I guess this 510 is my mean insert size.

so could I lie to simulator say: my mean insert size is 100! use:

--ismean 100 ?

what will be the consequence if i lie to simulator that the insert size is 100 ?

Many Thanks!!!!

[adamewing]

I suspect you're not going to wrangle BAMSurgeon into generating a realistic simulation of large deletions from exome data - that's unfortunately kind of out of scope for the assembly-based design of the program currently. Since any assembly will have gaps corresponding to the introns, modifying the entire region of the gene in one go won't be possible. Maybe have a look a tHapMix: https://github.com/Illumina/tHapMix

[Eric4Wu]

yeap, I found even I change --ismean & --issd will that BAMSurgeon stock in making mutation.

so I can't use BAMSurgeon to increase/decrease the reads in target exon region in normal bam file?

I will try tHapMix , thanks for reply! so BAMSurgeon is design for whole genome sequencing BAM file?

sorry for using WGS simulator for WES simulating use!