Dear Pier Francesco, 

actually the preferred sex is the one with different chromosomes, in order to dismiss the number of repeated sequences and for covering completely the genome of the species. If you sequence a female, XX, or a male ZZ, you're missing the other sex chromosome! whereas you're expending time and money into the sequencing of almost identical chromosomes (X or Z's!). Remember that one of the two sex chromosomes are "silenced" to keep the balance dose.

If you had a look in the already sequenced insect species, most of the specimens are males, meaning with this those with two different sex chromosomes.  

One point that you don't cover in your question is the nature of your insect. To have an almost perfectly covered genome, you should depart from a pure culture of your insect, or as pure as posible (nearly isogenic line). This will help during the assemblage and annotation of your genome altogether with the sequencing of a sample without redundancies (ie. those with two X chromosomes!). 

Have a look here i5K ... to check if your species is among the one selected, to see if someone can help you. 

https://www.hgsc.bcm.edu/i5k-pilot-project-summary

hope this helps you, 

Beatriz 

Dear Pier Francesco

Excellent suggestion by Beatriz. However, I would add that if you sequence XY(ZW) genome using high-throughput sequencing, you are likely to get only half of the reads for sex chromosomes compared to autosomes. This can pose significant challenges in assembly. Also, if X and Y share common regions of recombination, then it would be hard to assign those regions to either chromosomes owing to lower read depth etc.

In short, if you do plan to sequence XY(ZW) then you should consider the sequencing depth as well and possibly at least double it then planned for getting proper assembly of the sex chromosomes. Or else, you may miss out on both the X and the Y.

All the best for your project.

Hardip

Dear Hardip and Beatriz,

Thank you very much for you contributions. Actually, Beatriz, the question was posed mainly because of the issue that Hardip brought up: I cannot afford at the moment a great coverage (we're going to ~30x) and the fact that for sex chromosomes in males I will get only half the reads led me to think: should I get a female with no Y data but better coverage for the X, or a male? 

For what concerns the type of insect, it is a neuropteran, and the culture is near isogenic (the line coming from an F1 female, then all the subsequent generations come from inbreeding between the offspring at each generation). Unfortunately the species it is not between the i5K.

Another question: should this give me the opportunity to pool DNA from more individuals? 

I think that you need to have some idea about the condition of the sex chromosomes before making this decision. For example, mammalian sex chromosomes are highly repetitive and complex, and with the low coverage you describe and unless you use some large jumping libraries as well you are unlikely to get very large or useful scaffolds from the Y chromosome. However, if in your insect the sex chromosomes are relatively recent, then they will likely be easier to sequence. I would start by looking in the literature to find an estimate of the repeat content of the Y chromosome for your target species.

Another consideration is the reproductive state of the females. If you can get virgin females that’s fine. If you can get unfertilised eggs (haploid) that’s even better, as you get haploid information (gametic phase). Drosophila virgin female for example, will lay (unfertilized) eggs after a few days in isolation.  If the females are not virgin (fertilized), this may cause problem in your sequencing as you get reads from male sperm, in which case you may prefer sequencing males.