For me, intelligent design does not have to be called 'God.' A living cell has thousands of working parts that are animated. That does not just fall together in some primordial pond or under some rock. IMO, there has to be some intellect involved.
Well, so let's take a closer look at that.
According to the timetable, earth is about 4.5 billion years old. And, the first prokaryotic cells (mycoplasma, which is about the simplest cell we know of - it's basically a bacterium minus the cell wall) are somewhere around 4 billion years. So that leaves half a billion years for a primitive cell to form.
Mycoplasma has DNA, it's a single circular strand - which means, it requires tRNA. Transfer RNA can be considered an enzyme, but it's a small one, its chain length is only about 75 nucleotides. So let's be generous and use round numbers and say it requires 250 DNA base pairs to code for it, per type. Times 20 types, so about 5000 base pairs in total.
Mycoplasma also has 70s ribosomes, which contain about 4500 nucleotides, so let's say 15,000 base pairs. So for the native replicating machinery we're talking 20,000 base pairs. But let's be generous and call it 25k.
Next, mycoplasma requires about 50 different proteins to function and to self replicate. The total number of base pairs in a single strand of mycoplasma DNA is about 500,000, and the average sequence length of one of its proteins is about 400 amino acids, which would equate with 1200 base pairs, times 50, so about 60,000 base pairs in total.
So what do these numbers tell us? First, they tell us that mycoplasma has more DNA than it needs. By at least a factor of 5. What does the rest of the DNA do? No one really knows. But J Craig Venter and his colleagues in San Diego showed us that you can remove the excess and still have a fully functional self replicating mycoplasma. So the excess DNA is apparently "not necessary".
So here are two questions:
First, how long does it take to "evolve" a single strand of 500k DNA base pairs?
And second, if there is such an evolution, what are the chances of obtaining a working ribosome?
The first question is pretty easy to answer. Nucleotides can be synthesized directly and abiotically from acetaldehyde and glyceraldehyde-5-phosphate, the reaction takes about a day in a seawater solution at room temperature (the reaction is pH and temperature dependent, but the point is the raw material for DNA is very easy to come by). Then, once we have a supply of nucleotides we have to chain them together, and this is where we get our first glimpse of real evolution. Because nucleotide bonds form randomly but they are unstable, they break almost as fast as they make. To get something stable requires "nucleation", which means you need 4 or more base pairs in a row before water will stabilize the chain.
It turns out that this process is a lot easier to accomplish if you have RNA "first", which also means you have to have proteins "first". This is the origin of the RNA-world hypothesis. It says that instead of synthesizing the DNA abiotically, you do it from existing RNA. Which requires an enzyme called a polymerase, which is a protein. So before we can do this calculation, we need to do a similar calculation for proteins.
Scientists have done all these calculations. The short story is that half a billion years is a LONG time, you don't need nearly that much. You only need about 1/100 as much to get a full strand of 500,000 base pairs of DNA. And, this doesn't happen in just "one" micelle, it happens in billions upon billions of micelles simultaneously. Each one will have a slightly different result. So your chances overall, are the chances for one micelle, multiplied by a factor of a billion or more. The resulting answer is, it's not only possible, it's "very likely".
Which takes us immediately to the second question. If you have billions and billions of "tries", what are the chances that at least one of them will result in a working ribosome? The answer is, you get a huge advantage from the redundancy. If you stipulate that oceans cover half the earth, that gives you about 250 million square kilometers, so if you have one micelle per square meter you have a redundancy of about 250 trillion. Even without working through the detailed math you can see that the answer is once again "very likely". You have 250 trillion examples building at 1 extension per day and you only require 60,000. The odds are practically overwhelming, they approach 1.
We can now take this further by considering that the "failed" protein sequences can be reused for something else. Even if a protein fails as a polymerase, maybe it will succeed in moving phosphates around or linking glycine subunits or something. The picture that emerges is a vast collection of micelles with heterogeneous contents.
And now you have half a billion years to mix and match. Many of the micelles will get torn up on the rocks, where they will dump their contents back into the water. And where the loose protein will then be incorporated into other micelles by endocytosis.
The first gap in the evolutionary timeline is between cell stability and self replication, and now we see it isn't that much of a gap after all. A short time with this mix-and-match process will give you all kinds of synthase and reductase enzymes, only a few of which are actually required. The exchange of raw materials functions very much like a virus, it carries proteins and genetic information from one cell to another.
So the initial set of questions reduces to an issue of "how are the useful sequences kept around". Apparently a lot more is kept around than needs to be, because we have 5x as much DNA as we need.
If you actually think through this stuff you end up having to conclude it's eminently plausible. It doesn't require design, it just requires lots and lots of trial and error. Half a billion years worth.