It actually affects your cholesterol in every way pretty much - in a good way. It increases HDL, decreases LDL, decreases triglycerides, decreases overall cholesterol. It decreases inflammation and promotes nerve health, brain health, recovery. It reduces risk of heart disease, diabetes, even obesity due to the relationship of obesity and inflammation.
The evidence is rather mixed, actually. I don't know about basic science research, but in terms of actual clinical outcome results are conflicting. Overall, for both primary and secondary prevention of cardiovascular events, there appears to be no benefit (or rather, a non-significant benefit) to taking omega-3. It may be that future, larger, more rigorous studies will allow subsequent systematic reviews to show an overall benefit; but then again, it may not.* The jury's still out.** However, given the good side-effect profile and tolerance, many studies advocate against not taking omega-3 supplements, under the provision that even if it does no good, at least it does no harm. I respectfully disagree; it costs money, and because of its possible absence of effect, it may cause people to indulge in more risky behavior; similar to how application of sunscreen, no matter how poorly, causes people to engage in longer and more dangerous periods of sun exposure in the mistaken belief that they are protected somehow.
If you feel an overwhelming need to do something about reducing cardiovascular risk, quitting smoking, eating a healthier diet, and exercising more are all preferable; heck, even taking a statin may be better for the same indication; at least it is shown to have significant effect, even in primary care, and it may even be cheaper.
*If there is a benefit, it is miniscule. Regardless of how omega-3 appears to influence cellular function and inflammation, in clinical practice effect, when it is there, is rarely more than a few percentage points better than placebo (if you call olive oil a placebo). Compare this to the effect of statins, which show a 20-50% improvement in lipid profile and morbidity/mortality, depending on the dose and type of statin used.
**Note that there is an element of bias to this view; if future studies show a statistically significant effect, then the case is settled, according to all observers. However, if future studies fail to show a statistically significant effect, then there will be perpetual calls for more research in the hopes of elucidating a previously-hidden positive effect.