Deep learning takes on protein folding: In 1972, Anfinsen won a Nobel prize for demonstrating a connection between a protein’s amino acid sequence and its three-dimensional structure. Since 1994, scientists have competed in the biannual Critical Assessment of Structure Prediction (CASP) protein-folding challenge. Deep learning methods took center stage at CASP14, with DeepMind’s Alphafold2 achieving remarkable accuracy. Baek et al . explored network architectures based on the DeepMind framework. They used a three-track network to process sequence, distance, and coordinate information simultaneously and achieved accuracies approaching those of DeepMind. The method, RoseTTA fold, can solve challenging x-ray crystallography and cryo–electron microscopy modeling problems and generate accurate models of protein-protein complexes. —VV

Reductions play a key role in organic synthesis, producing chiral products with new functionalities. Enzymes can catalyse such reactions with exquisite stereo‐, regio‐ and chemoselectivity, leading the way to alternative shorter classical synthetic routes towards not only high‐added‐value compounds but also bulk chemicals. In this review we describe the synthetic state‐of‐the‐art and potential of enzymes that catalyse reductions, ranging from carbonyl, enone and aromatic reductions to reductive aminations.

A series of synthetic nicotinamide cofactors were synthesized to replace natural nicotinamide cofactors and promote enoate reductase (ER) catalyzed reactions without compromising the activity or stereoselectivity of the bioreduction process. Conversions and enantioselectivities of >99% were obtained for C═C bioreductions, and the process was successfully upscaled. Furthermore, high chemoselectivity was observed when employing these nicotinamide cofactor mimics (mNADs) with crude extracts in ER-catalyzed reactions.

Bacteria can reduce toxic and carcinogenic Cr(VI) to insoluble and less toxic Cr(III). Thermus scotoductus SA-01, a South African gold mine isolate, has been shown to be able to reduce a variety of metals, including Cr(VI). Here we report the purification to homogeneity and characterization of a novel chromate reductase. The oxidoreductase is a homodimeric protein, with a monomer molecular mass of approximately 36 kDa, containing a noncovalently bound flavin mononucleotide cofactor. The chromate reductase is optimally active at a pH of 6.3 and at 65°C and requires Ca ²⁺ or Mg ²⁺ for activity. Enzyme activity was also dependent on NADH or NADPH, with a preference for NADPH, coupling the oxidation of approximately 2 and 1.5 mol NAD(P)H to the reduction of 1 mol Cr(VI) under aerobic and anaerobic conditions, respectively. The K _m values for Cr(VI) reduction were 3.5 and 8.4 μM for utilizing NADH and NADPH as electron donors, respectively, with corresponding V _max values of 6.2 and 16.0 μmol min ⁻¹ mg ⁻¹ . The catalytic efficiency ( k _cat / K _m ) of chromate reduction was 1.14 × 10 ⁶ M ⁻¹ s ⁻¹ , which was >50-fold more efficient than that of the quinone reductases and >180-fold more efficient than that of the nitroreductases able to reduce Cr(VI). The chromate reductase was identified to be encoded by an open reading frame of 1,050 bp, encoding a single protein of 38 kDa under the regulation of an Escherichia coli σ ⁷⁰ -like promoter. Sequence analysis shows the chromate reductase to be related to the old yellow enzyme family, in particular the xenobiotic reductases involved in the oxidative stress response.