Naive question

[Sthira]

 

Naive question is this: if scientists can create substances that do so much bodily harm that they "...cause 150 mutations in every lung cell..." then why can't scientists create substances that do equal measures of bodily good, for example, like protect and repair bodies from aging damage?

 

Maybe you saw this story about, yawn, how bad cigarettes are for poor, addicted humans. Ok fine we get it, but where is the lab-created opposite of a cigarette, something that protects and repairs human bodies?

 

Why is it so easy to screw ourselves up -- here, have a toke -- yet so difficult when we want to do good?

 

 

Smoking a pack a day causes 150 mutations in every lung cell, research shows

 

LONDON—Scientists have found that smoking a pack a day of cigarettes can cause 150 damaging changes to a smoker's lung cells each year.

 

The findings come from a study of the devastating genetic damage, or mutations, caused by smoking in various organs in the body.

 

Publishing in the journal Science on Thursday, Nov. 3, the researchers said the findings show a direct link between the number of cigarettes smoked in a lifetime and the number of mutations in the DNA of cancerous tumors.

 

The highest mutation rates were seen in lung cancers, but tumors in other parts of the body—including the bladder, liver and throat—also had smoking-associated mutations, they said. This explains why smoking also causes many other types of cancer beside lung cancer.

 

Smoking kills six million people a year worldwide and, if current trends continue, the World Health Organization predicts more than 1 billion tobacco-related deaths this century.

 

Cancer is caused by mutations in the DNA of a cell. Smoking has been linked with at least 17 types of cancer, but until now scientists were not clear on the mechanisms behind many of them.

Ludmil Alexandrov of Los Alamos National Laboratory in the United States, one of those who carried out the research, explained that in particular, it had until now been difficult to explain how smoking increases the risk of cancer in parts of the body that don't come into direct contact with smoke.

 

"Before now, we had a large body of epidemiological evidence linking smoking with cancer, but now we can actually observe and quantify the molecular changes in the DNA," he said.

 

This study analyzed over 5,000 tumors, comparing cancers from smokers with those from people who had never smoked.

It found certain molecular fingerprints of DNA damage—called mutational signatures—in the smokers' DNA, and the scientists counted how many of these were in different tumors.

 

In lung cells, they found that on average, smoking a pack of cigarettes a day led to 150 mutations in each cell every year. Each mutation is a potential start point for a "cascade of genetic damage" that can eventually lead to cancer, they said.

 

The results also showed that a smoking a pack of cigarettes a day led to an average 97 mutations in each cell in the larynx, 39 mutations for the pharynx, 23 for the mouth, 18 for the bladder, and six mutations in every cell of the liver each year.

 

Mike Stratton, who co-led the work at Britain's Wellcome Trust Sanger Institute, said it was a bit like digging in to the archaeology of each tumor.

 

"The genome of every cancer provides a kind of archaeological record, written in the DNA code itself, of the exposures that caused the mutations," he said. "Looking in the DNA of cancers can provide provocative new clues to how (they) develop and thus, potentially, how they can be prevented."

[Dean Pomerleau]

Sthira,

 

Why is it so easy to screw ourselves up -- here, have a toke -- yet so difficult when we want to do good?

 

Short answer - entropy. There are a lot more ways to screw up a complex system than to improve it. Consider your iPhone (or Android phone). Open it up. Take out a screwdriver. Try to improve the phone's performance by poking around with its internal components with said screwdriver. Good luck.

 

BTW, same goes for the delicate structure of a complex human society. There are many more ways to tear it down than to build it up. Vote.

 

--Dean

[Sthira]

So where are the alternate simulations -- live experiments that are not spent to exhaustion succumbing to entropy, but rather are improved upon with passing days?

 

Where is a cigarette's opposite?

[Dean Pomerleau]

So where are the alternate simulations -- live experiments that are not spent to exhaustion succumbing to entropy, but rather are improved upon with passing days?

 

Where is a cigarette's opposite?

 

Hmmm... Maybe you are thinking of the process of evolution, where tweaks that work are retained. It's been happening for hundreds of millions of years, but is a very slow, incremental process. For example, after 5K years since the mutation was discovered, a substantial fraction of us can now digest lactose as adults, since it helped ancestors in northern climates survive and reproduce when other sources of food were scarce in winter.  But evolution doesn't care (much) about longevity or long-term health (possible minor and controversial exceptions - the "grandmother effect"). If it doesn't help us produce more offspring, evolution isn't going to favor it.

 

--Dean

[Dean Pomerleau]

Interesting coincidence. Immediately after writing the above post, this new paper [1] came across my (twitter) radar. It suggests AI and deep neural networks may have a role to play in searching for genetic codes that promote the expression of other genes. It's a long way to go from your dream of AI solving the problems of human illness and death, but it's a step in that direction.

 

--Dean

 

-------------

[1] 

Predicting Enhancer-Promoter Interaction from Genomic Sequence with Deep Neural Networks

Shashank Singh, Yang Yang, Barnabas Poczos, Jian Ma

doi: http://dx.doi.org/10.1101/085241

This article is a preprint and has not been peer-reviewed [what does this mean?].

 

• Abstract
• Info/History
• Metrics
•  Preview PDF

 

Abstract

In the human genome, distal enhancers are involved in regulating target genes through proximal promoters by forming enhancer-promoter interactions. However, although recently developed high-throughput experimental approaches have allowed us to recognize potential enhancer-promoter interactions genome-wide, it is still largely unknown whether there are sequence-level instructions encoded in our genome that help govern such interactions. Here we report a new computational method (named "SPEID") using deep learning models to predict enhancer-promoter interactions based on sequence-based features only, when the locations of putative enhancers and promoters in a particular cell type are given. Our results across six different cell types demonstrate that SPEID is effective in predicting enhancer-promoter interactions as compared to state-of-the-art methods that use non-sequence features from functional genomic signals. This work shows for the first time that sequence-based features alone can reliably predict enhancer-promoter interactions genome-wide, which provides important insights into the sequence determinants for long-range gene regulation.

[Sthira]

Ok. I accept that. What I hear you saying is that the world sucks, it was simulated to be an entropy experiment, and here we are in free fall. We may easily screw ourselves up with anything -- even too much kale and meditation -- but to improve ourselves is just really, really hard because it's all so complicated. The fact that life is complex, however, takes a backseat to the fact that it's falling apart.

[Dean Pomerleau]

Sthira,

 

Whether or not my hypothesis about our world being a simulation is correct, it's hard to imagine a world without entropy. In fact, some tendency for things to "fall apart" is probably necessary for evolution to happen in the first place. If nothing fell apart, it would be one big amorphous clump, and so little bits couldn't get rearranged in interesting new combinations so evolution couldn't operate.

 

"Creative destruction" is how the world operates, whether it is a simulation or not.

 

--Dean

[Sthira]

But why? Why not create a simulation that works in reverse? Instead of things falling apart, things start from a sad state of brokenness and get better and better?

[Dean Pomerleau]

Sthira,

 

But why? Why not create a simulation that works in reverse? Instead of things falling apart, things start from a sad state of brokenness and get better and better? 

 

That's exactly what is happening, through the process of evolution. But it is a huge (to put it mildly) space to search, so progress is slow. With humans, and now CRISPR, the search has sped up dramatically in recent millennia through the process of cultural evolution, and now targeted scientific investigations into health, genetics, etc. AI promises to accelerate the search for better solution even further, whether or not we remain in our meat-bag bodies. 

 

If you're asking why not skip the slow and messy process of evolution, and instead simply jump to the "right" answer, or only explore parts of the search space where "good" answers lie, that would take an omniscient creator of all this, and I don't think such an entity exists. Remember, my hypothesis is that the creators of our (simulated) world are not that much more advanced than we are, at most a 100 or 200 years in our future, and probably much less. So they are neither omniscient, omnipotent nor (certainly) omni-benevolent.

 

--Dean

[Sthira]

If you're asking why not skip the slow and messy process of evolution, and instead simply jump to the "right" answer, or only explore parts of the search space where "good" answers lie, that would take an omniscient creator of all this, and I don't think such an entity exists. Remember, my hypothesis is that the creators of our (simulated) world are not that much more advanced than we are, at most a 100 or 200 years in our future, and probably much less. So they are neither omniscient, omnipotent nor (certainly) omni-benevolent.

 

--Dean

If AGI is progressing as rapidly as we've (and especially you've) been suggesting, then that AGI is to become the omniscient, omnipotent, and hopefully omni-sweet to us (you know, like we're so good and kind to all the plants and animals we share the planet with, he said with sarcasm stuck between molars).