Outgroups are used to root phylogenies. This allows you to polarize characters. Alternatively you would have to place the root at the midpoint of your unrooted tree.

Dear Niranjan

If your purpose is to determine the evolutionary relationships of the ingroup (test species), it is highly recommended to have an outgroup (the reference group). An outgroup can give you a sense of where on the bigger tree of life the main group of organisms falls (https://evolution.berkeley.edu/evolibrary/article/phylogenetics_02)

You can also read the relevant answers in the following link:

https://www.researchgate.net/post/How_can_you_choose_the_best_outgroup_for_phylogenetic_analysis

Thank you dear Jason about root of phylogeny, is it correct to keep a close genus when we compare species level?

Nice inputs given by Sunil ji, further inputs are well come,

In some programs you may be compelled to use a single outgroup, but it is often good to include multiple, with a 'far out' group to root the whole tree. Phylogenetic results can be affected by the taxa you include (or exclude). I would recommend having as many representatives of close genera that you can manage and a few less closely related taxa to help root the tree. As an example of using lots of outgroups see: Debevec, AH., Cardinal, S & Danforth, BN. (2012). Identifying the sister group to the

bees: a molecular phylogeny of Aculeata with an emphasis on the superfamily Apoidea.

—Zoologica Scripta, 41, 527–535.

Yah,

What would be the reson some times a genus clading middle of another single genera?

Single genus?