There are clearly other biological hypotheses that have somehow graduated to the orthodoxy, such as Earthly abiogenesis. The hypothesis that non-life to life transition happened on Earth is a reasonably testable one, but a lot more mileage is given to experiments that assume early Earth origins (ie. that somehow Earthly origin is almost certainly the "truth"). However, I will concentrate on those biases against anything that remotely appears Lamarckian or like intelligent design (even if the intelligence is held within the DNA and/or the organism itself).
Now my thesis is that although the main feedback of information back to the DNA of how well it is doing is via the brute force of "natural selection" - that is a comparative between individuals with different DNA. However, my thesis is that there clearly is processes that are very much like selection that is also giving a feedback of how "fit" the DNA is. There are tens to hundreds of millions of sperm that are competing for an egg. The male is producing these and there may be a selection process from that side which strengthens those sperm with particular mutations and weakens others depending in part on the level and type of stress he is under. On the female side, the same thing is bound to be happening, with body conditions slowing down sperm with some particular mutations and letting through others depending on the level and type of stress she is under. As a final error correcting check, particular important sequences of genes are tested in the controlled conditions of the ova. The ova has been fully formed since before the mother-to-be was born. This is a rudimentary "archive" that checks on crucial DNA sequences to make sure there is no changes to those. Thus, a great deal of crucial environmental and/or competitive information is fed back to the DNA, as well as strict error correction on crucial segments during and even before conception. Thus the new organism will have had extensive pre-birth selection to give its genes the best chance of survival.
This might yet be uncontroversial - But it really depends on how sophisticated the stress to mutation-selection link is. To me it is plausible that there is a simulation engine powered by the subconscious mind that does intense calculations on how well any particular mutations may benefit particular constraints. Information can flow through the conscious mind, seeing certain genetic consequences, and the subconscious could translate it to stress information that would be selective for sperm proxies of the same genetic consequences. These would be really, really smart mutations - Much smarter than any genetic engineer could ever hope to become in following millenia.
The difference between the orthodox view and my view can be illustrated by a fashion paradigm. In the orthodox view, fashion designers make random changes to existing designs and force models to wear them. If they sell, they make more of the same. If they don't, the clothes get thrown in the bin and they don't make anymore of that crap.
In my view, what (successful) fashion designers do is that fashion designers make random changes to existing designs - show them to a test audience (of horny males, apparently :)) - changes that show some approval get kept. Process is repeated as many times as the exhausted sewing machinists and test models can cope with and then the surviving clothing is forced onto the model and onto the catwalk.
In my more sophisticated fashion design view, the machinists and models are given a break and all the experimentation with random changes happen on the computer (or a drawing board) and get shown to the test audience. By the time the design gets on the catwalk, the design looks "worked" and unlike the designs it may have been based on. To an untrained eye, it might still look like random changes, but those that *actually* put random changes straight on to the catwalk don't win the fashion show.
Additionally - to look at it another way the *winner* of the fashion show can gloat and make grandiose claims about his *creativity*, when in truth he just had a more sophisticated system of randomised changes with good feedback loop.