Most medicines trace back to something found in nature: a plant, a fungus, a sea creature, or a mineral pulled from the earth. Roughly half of all drugs approved over the last 30 years come directly or indirectly from natural products. The other half are designed in laboratories, but even many synthetic drugs started as copies or modifications of molecules that scientists first discovered in living organisms. The journey from raw source to pharmacy shelf typically takes more than 10 years, and only about 19% of compounds that enter human testing ever reach approval.
Plants: The Oldest Medicine Cabinet
For most of human history, plants were medicine. Bark, leaves, roots, and berries were brewed, chewed, or ground into poultices long before anyone understood the chemistry behind them. That tradition wasn’t guesswork. Of 122 plant-derived compounds used as drugs worldwide today, 80% have a traditional medicinal use that matches the drug’s current purpose. Indigenous healers, in other words, had already figured out what many of these plants were good for.
The most iconic example is aspirin. In 1899, Bayer introduced a purified version of a compound originally isolated from white willow bark. It became the first semi-synthetic drug, meaning chemists took a natural molecule and tweaked it in the lab to make it safer and more effective. That same pattern repeated throughout the 20th century. The cancer drug paclitaxel (sold as Taxol) was discovered in the bark of the Pacific yew tree and is now used to treat lung, ovarian, and breast cancer. Demand runs between 100 and 200 kilograms per year, enough for roughly 50,000 treatments, so it’s now produced synthetically rather than harvested from trees.
Morphine, codeine, and even heroin all trace back to the opium poppy. In the 1870s, chemists boiled crude morphine with an acid to create diacetylmorphine (heroin) and found it could be converted into codeine, a painkiller still widely prescribed. These early experiments with plant chemistry laid the groundwork for the entire pharmaceutical industry.
Fungi and Soil Bacteria
The ground beneath your feet is one of the richest sources of medicine ever discovered. Over 5,000 antibiotics have been identified from bacteria and fungi cultured from soil, though only about 100 have made it into commercial use. The most famous is penicillin, discovered by Alexander Fleming in 1929 growing on a mold called Penicillium notatum. It took another decade for researchers to figure out how to produce it at scale, but once they did, synthetic penicillins transformed medicine and launched the antibiotic era.
The majority of antibiotics come from a group of soil-dwelling bacteria called Streptomycetes. These microorganisms produce a staggering variety of biologically active compounds. In total, about 23,000 secondary metabolites (chemicals organisms make that aren’t essential for their own survival but often have potent effects on other organisms) have been cataloged. Roughly 42% come from bacteria in the Actinobacteria group, another 42% from fungi like Penicillium species, and the remaining 16% from other bacteria.
Animal Venoms
Some of the most effective medicines started as poisons. The Brazilian pit viper produces venom containing peptides that cause a dangerous drop in blood pressure in its prey. In the 1970s, researchers figured out exactly how those peptides worked: they block an enzyme that tightens blood vessels. That insight led to captopril, the first in a class of blood pressure medications called ACE inhibitors, now one of the most widely prescribed drug families in the world for treating hypertension and heart failure.
The Gila monster, a venomous lizard native to the American Southwest, contributed to diabetes treatment. Its saliva contains a compound that mimics a hormone involved in blood sugar regulation. A synthetic version, sold as Byetta, became an FDA-approved treatment for type 2 diabetes and helped pave the way for newer weight-loss medications in the same drug class.
The Ocean Floor
Marine organisms are relative newcomers to drug discovery, but they’ve already produced 13 approved drugs in the U.S. and Europe, with four approved in just the last three years. Ten of those 13 are cancer treatments.
Sea sponges have been particularly productive. In the early 1950s, researchers isolated unusual compounds from a Caribbean sponge, which eventually led to cytarabine, a leukemia drug that entered the market in 1969. A Japanese sponge yielded the compound behind eribulin, now used for breast cancer and a rare type of soft-tissue cancer. Cone snails, slow-moving predators that paralyze fish with venom, provided the basis for ziconotide, a powerful chronic pain medication. And tunicates, small filter-feeding animals that attach to rocks and ship hulls, are the source of trabectedin, used against ovarian cancer and soft-tissue sarcomas.
Minerals and Metals
Not all medicine comes from living things. Minerals and metals have been used therapeutically for centuries, and several remain important today. Lithium, a simple element found in rocks and mineral springs, is a cornerstone treatment for bipolar disorder. Platinum-based compounds are among the most effective chemotherapy drugs for solid tumors. Aluminum compounds have been added to billions of vaccine doses over the past 90 years to boost immune response.
Bismuth, the active ingredient in common stomach remedies, also has antimicrobial properties and can kill the bacterium responsible for most stomach ulcers. Arsenic compounds, used in traditional Chinese medicine for centuries, have found modern clinical applications as cancer treatments. Even lanthanum, a rare earth element, is now an established drug for removing excess phosphate from the body in people with chronic kidney failure.
Biologics: Grown in Living Cells
A growing share of modern medicines aren’t extracted from nature or built through chemistry. They’re grown. Biologic drugs, including monoclonal antibodies used to treat cancer and autoimmune diseases, are manufactured inside living mammalian cells. The cells are genetically engineered to produce a specific therapeutic protein, then cultivated in large bioreactors where they multiply and churn out the drug.
Chinese hamster ovary cells produce 60 to 70% of all recombinant biologics on the market. Mammalian cells are preferred over bacteria or yeast because they can fold and assemble complex proteins correctly, adding the sugar-based modifications that many drugs need to function in the human body. This approach powered early cancer breakthroughs like Rituxan (1997) and Herceptin (1998) and now underpins a large portion of the pharmaceutical industry.
Synthetic and AI-Designed Drugs
Starting in the 1980s, researchers hoped that combinatorial chemistry, a method of rapidly generating thousands of new molecular structures in the lab, would replace natural product discovery. That largely didn’t happen. In the decades since, only one drug created purely through combinatorial chemistry has received FDA approval: a kidney cancer treatment approved in 2005.
What has changed the landscape is the combination of molecular biology and, more recently, artificial intelligence. AI-designed therapeutics are now in human trials across multiple disease areas. Generative chemistry platforms can propose entirely new molecular structures, while self-driving laboratories use robotics to rapidly build and test those molecules. These systems don’t replace the natural world as a source of drug inspiration, but they dramatically compress the timeline for turning a promising molecule into a viable treatment candidate.
How Long the Journey Takes
Regardless of where a drug originates, getting it from initial discovery to your pharmacy is a slow, expensive process that typically spans more than a decade. A compound first goes through preclinical testing in the lab, where only about 32% of candidates survive. Those that pass enter three phases of human trials. About 75% make it through Phase I (basic safety testing in small groups), 50% survive Phase II (testing whether the drug actually works), and 59% clear Phase III (large-scale trials confirming effectiveness and monitoring side effects). Even after all that, roughly 12% fail at the final regulatory review stage.
From the moment a compound first enters human testing, the overall probability of reaching approval is about 19%, or roughly one in five. That low success rate, combined with the years of research required at each stage, is a major reason why drug development is so costly and why the search for new sources of medicine, from deep-sea organisms to AI-generated molecules, never stops.