A single eukaryotic mRNA is translated to yield a typical-sized protein in about 30–60 seconds. During this process, translation accuracy is maintained by elongation factors, EF-Tu and EF-G in bacteria, and EF1 and EF2 in eukaryotes. During translation, EF-Tu associates with GTP and aminoacyl-tRNA. Together, they bind the A site of the ribosome to form a codon-anticodon duplex. Each codon–anticodon interaction is checked twice. The first checkpoint is the complementary base pairing between the mRNA codon and the tRNA anticodon. A correct codon–anticodon match binds more tightly than an incorrect tRNA, which will dissociate. At the second checkpoint, part of the 16S rRNA in the small ribosomal subunit folds around the paired bases forming a network of ribosomal contacts that are specific for each position of the codon–anticodon. A mismatched tRNA fails to make ribosomal contacts, and dissociates, while the correct tRNA changes the conformation of the catalytic center within the ribosome. This is called an induced fit. This causes EF-Tu to hydrolyze GTP and dissociate from the ribosome. The released aminoacyl-tRNA can now participate in translation. If an aminoacyl-tRNA escapes proofreading and is added to the growing polypeptide chain, a codon‒anticodon mismatch in the P site of the ribosome causes an increased rate of error in the A site. With successive rounds of amino acid misincorporation, the faulty polypeptide chain is prematurely terminated by release factors and the flawed protein is degraded.