Personality traits and temperament are very visibly inherited (to a level of granularity that can be quite astonishing), leaving the question of causal mechanism for that inheritance: the nature/ nurture balance (plus mechanism detail and causes of variation).This study follows in the path of past twin studies looking at even narrower outcomes (notably IQ, which is a little reductive but has relatively well defined measurement, and where inheritance is almost entirely genetic). But this is still only tickling the surface.
Anecdotal: having both biological siblings and an adopted sibling in our family from a few months of age (and knowing the biological father of the adopted sibling), it seems very obvious to me that the main mechanism of personality trait inheritance is genetic. (Assuming no calamitous abuse, trauma, malnutrition, childhood illness, obesity, substance abuse... or similarly devastating lifestyle patterns that can propagate across generations and do impact personality.)
Some more context for setting sensible priors/ expectations: humans are just animals. We recognise personality traits and behaviour patterns in particular dog breeds (and of course all domesticated animals) - we actively breed for such traits. We know that these traits persist and continue to be inherited irrespective of animal parenting. It would be absurd to expect that genetics aren't a large part of the explanation for variation in personality traits and behaviour patterns within human populations; and also for their heritability and for variation across distinct human populations with historic barriers to gene flow.
What's missing, of course, is more rigorous and more granular definition of traits - and then measurement, and data collection.
We're also missing a more direct attribution to the underlying genetics (sequenced genomes, activations...) - though data there ought to become accessible.
Of course, low level personality traits influence behaviour, habits and decisions in a manner mediated by our environment, nutrition, activity, stressors, hormonal fluctuations, social institutions, narratives, learned perception and learned actions/ skills (where patterns of thought are a type of skill too!).
There's my outline inference, reasoning & expectations. It would be nice to see a framework of more rigorous definitions, data collection channels and falsifiable hypotheses - to more robustly expand our understanding of human nature(s).
With all knowledge, there is the question of how it might change the world. The ability to make better personality or (limited) behavioural predictions about a person based on genetics, could be profitably (ab)used in setting insurance premiums/ deciding insurance eligibility, in dating/ partner selection, in recruitment, in provision of better (psychological) healthcare, in providing better (and better individualized) education, in providing better (customized) advice for people to improve their lifestyles, habits and outcomes, in selective IVF, in the migration policy of nation states or in some sort of active eugenics program. Many of those would seem ethically problematic! That's perhaps for democracy & legislation to address - later, once patterns and mechanisms are better explored and modeled.
Separately from the practical applications, a better understanding is in itself good - we should want to understand ourselves; we should seek the truth. It's a shame this sort of research is (still) so controversial. Tooling progress (especially sequencing and neural networks, where neural networks have a role in trait discovery/ measurement and in inference) will most probably drive this field forwards in coming decades, irrespective of controversy.