Insertional mutagenesis is exactly what it sounds like; inserting DNA to mutate the organism (that is, change its genetic code).

What don't you understand about genetic footprinting?

A couple of things first:

The sequencing a gene does not immediately mean that we UNDERSTAND THE FUNCTION.

Open Reading Frames ( the ORF in your pdf link ) are just sections of a sequence that COULD code for a gene. You'd look for a START codon, then followed by a variable number of nucleotide triplets, and finally a STOP codon. The longer the sequence, the higher the chance there is a gene located around it.

Now that you know approximately where it is, you can start inserting KNOCKOUTS into the segment of interest, hopefully deleting a function gene and then you would ( in theory ) be able to better understand what segment X does or doesn't do in the organism.

I like your question: If one sees a trait in an organism, how can one find the genes involved?

Not having done this myself here are my 2c:

First I'd try to see if the genes for the trait had already been mapped, I'd check scientific journals and patents. The less I have to do myself the better.

Like you said if one has an idea of what genes are involved in a trait it's possible to break them(knock them out) and see if it affects the trait.

In eukaryotes the position of a gene can be approximated thanks to genetic linkage.

If it's a protein coding gene and the protein can be purified there are methods to sequence it. The codons can be mapped against sequenced DNA.

Genetic foot printing means mapping many traits to genes at once. As I understand it for every trait there needs to be a selection method that will kill off the cells not having the trait. The sequence of the strains surviving the selections are compared to each other as well as to the parent mutated gene pool.

I don't know how useful it is to compare sequences after a single selection(base vs mutated and selected) for an unmapped species. For one thing the selected sequence will at least also contain the chassis(the bare essentials) for the organism which somehow has to be separated from the genes of interest.

There are additional methods for different circumstances. A well known one is attaching GFP to the protein coding gene of interest.

If one finds the genes but they are for a different species they might or might not work in the chosen species. If it they don't work because of a non trivial reason it can be difficult to understand why(there is a paper in it for you if you are a scientist though). There are known tweaks one can try such as changing codons.

This is a personal opinion: The less work that needs to be done the better. Chances are someone has already mapped the trait and then that's one less thing to deal with. I think most would be biohackers don't want to spend time inventing and mapping out what is essentially infrastructure. The higher the abstraction level and the more that can be accomplished in less time the better.

Gene mapping is one of those things companies and organizations do well, it scales up well. The biohackers job is to be creative with the data.

cu

-Splicer