Humans Have Only 20,500 Protein-Encoding Genes

The first drafts of the human genome indicated about 30,000 genes, a number that was very much in line with many predictions that had been made over the years by scientists who were studying the topic. (Other scientists, and most science writers, thought there were about 100,000 genes [Facts and Myths Concerning the Historical Estimates of the Number of Genes in the Human Genome]).

Since the publication of the first draft, the number of genes has been dropping as annotators eliminate sequences that were falsely attributed to protein-encoding genes. Current estimates suggest there are about 28,000 different genes all together with about 4,000 of them encoding RNA products such as ribosomal RNA, tRNA, and the small RNAs involved in a numer of metabolic processes [Ensembl: Homo sapiens].

A gene encoding a protein will have an open reading frame (ORF) consisting of multiple codons— usually more than 100. Some of these potential protein-encoding genes appear to be unique to humans. They weren't found in the other mammalian genomes that had been sequenced (e.g., mouse, dog). Quite a few scientists took this as evidence for genes that distinguish humans from other mammals. According to them, these unique genes arose during the recent evolution of Homo sapiens and that's why there are no homologues in the other mammalian genomes.

Other scientists looked at the data in a different light. They suspected that these "unique" or "orphan" genes were more likely to be artifacts because they were not conserved. In other words, they reached exactly the opposite conclusion based on their understanding of evolution. Their prediction was that these orphan genes resulted from spurious ORF's and not real genes.

This problem has been examined by Eric Lander's group in Boston, MA (USA) and the results were published in PNAS (Clamp et al., 2007). Their careful analysis has eliminated most of the orphan genes and the new gene count for protein-encoding genes is now 20,488.

Here's how the authors describe the purpose of their study,

The purpose of this article is to test whether the nonconserved human ORFs represent bona fide human protein-coding genes or whether they are simply spurious occurrences in cDNAs. Although it is broadly accepted that ORFs with strong cross-species conservation to mouse or dog are valid protein-coding genes (7), no work has addressed the crucial issue of whether nonconserved human ORFs are invalid. Specifically, one must reject the alternative hypothesis that the nonconserved ORFs represent (i) ancestral genes that are present in our common mammalian ancestor but were lost in mouse and dog or (ii) novel genes that arose in the human lineage after divergence from mouse and dog.

To begin the study they choose to analyze the 21,895 protein-encoding genes in the Ensembl database. They looked for genes that were related to similar sequences in the mouse and dog genomes. (These are the only two well-characterized non-human, mammalian genomes.) After visual inspection of low scoring sequences they were able to eliminate about 1600 potential genes because they were pseudogenes, transposons, or artifacts of various sorts.

They were left with 19,108 verified genes and 1177 orphan "genes"—human ORF's that were not similar to any gene in the mouse and dog genomes. These genes could be newly evolved genes in the human/primate lineage or ancient genes that had been lost in mice and dogs.

The next step was to categorize the orphan "genes" to see if they looked like real protein-encoding genes. The results indicated that in terms of sequence similarity to the same regions in the mouse and dog genomes, the orphan ORF's were indistinguishable from random sequences. Similarly, the characteristics of the presumed codons of these genes were very different from conserved genes and very similar to random sequences with short accidental reading frames. Thus, the orphan sequences look like artifacts.

To confirm this conclusion, the authors compared the sequences to the macaque and chimpanzee genomes. They were not found in those genomes either.

If the orphans represent valid human protein-coding genes, we would have to conclude that the vast majority of the orphans were born after the divergence from chimpanzee. Such a model would require a prodigious rate of gene birth in mammalian lineages and a ferocious rate of gene death erasing the huge number of genes born before the divergence from chimpanzee. We reject such a model as wholly implausible. We thus conclude that the vast majority of orphans are simply randomly occurring ORFs that do not represent protein-coding genes.

This analysis was extended to the other gene catalogs (Vega, and RefSeq) as well as an updated version of the Ensembl catalog (v38). This resulted identification of an additional 1271 valid genes. Adding in the genes in the mitochondrial genome (13) and the Y chromosome (78) gives a total of 20,470 genes.

Finally, reanalysis of the transposons and pseudogenes revealed 18 cases where a real gene had evolved from an inactive pseudogene. This gives a grand total of 20,488 protein-encoding genes in the human genome.

There are several conclusions that can be drawn from this excellent study.

We show that the vast majority of ORFs without cross-species counterparts are simply random occurrences. The exceptions appear to represent a sufficiently small fraction that the best course is would be consider such ORFs as noncoding in the absence of direct experimental evidence.

This is going to be a major challenge for many workers who prefer to see evolution in a different manner. There are a number of papers that view these orphans sequences as direct evidence that human specific genes had arisen in the recent past. Clamp et al. (2007) are saying that if the sequences aren't present in the macaque and chimpanzee then one should conclude that they are artifacts.

Remember, many of the artifactual genes are supported by EST/cDNA data suggesting that they are transcribed. This study calls that evidence into question—correctly in my opinion—indicating that we should be skeptical of the EST data.

One important biological implication of our results is that truly novel protein-coding genes (encoding at least 100 amino acids) arise only rarely in mammalian lineages. With the current gene catalogs, there are only 168 "human-specific" genes (<1% of the total; only 11 are manually reviewed entries in RefSeq; see SI Table 4). These genes lack clear orthologs or paralogs in mouse and dog, but are recognizable because they belong to small paralogous families within the human genome (2 to 9 members) or contain Pfam domains homologous to other proteins. These paralogous families shows a range of nucleotide identities, consistent with their having arisen over the course of ~75 million years since the divergence from the mouse lineage.

This is an important conclusion and I think it is accurate. There are very few "new" genes in the human genome, and, by implication, in other mammalian genomes. This conclusion is consistent with what we know about evolution but it contradicts studies that purport to show rapid evolution of novel genes and novel regulatory mechanisms in humans.

---

[Image Credit: The human karyotype is from the Ensembl website.]

Clamp, M., Fry, B., Kamal, M., Xie, X., Cuff, J., Lin, M.F., Kellis, M., Lindblad-Toh, K. and Lander, E.S. (2007) Distinguishing protein-coding and noncoding genes in the human genome. Proc. Natl. Acad. Sci. (USA) 104:19428-19433. [DOI 10.1073/pnas.0709013104]