China
Africa
Explore chapters and articles related to this topic
Reliable Biomedical Applications Using AI Models
Published in Punit Gupta, Dinesh Kumar Saini, Rohit Verma, Healthcare Solutions Using Machine Learning and Informatics, 2023
Shambhavi Mishra, Tanveer Ahmed, Vipul Mishra
Obtaining the best modeling knowledge of the protein folding process is currently one of the most difficult topics in omics [78]. During the gene expression process, each protein fold goes into a special type of 3-D structure which specifies its biological functionalities. The majority of these studies are focused on predicting protein secondary and tertiary structures.
Microbial Pathways of Lipid A Biosynthesis
Published in Helmut Brade, Steven M. Opal, Stefanie N. Vogel, David C. Morrison, Endotoxin in Health and Disease, 2020
Paul D. Rick, Christian R. H. Raetz
Lipid A synthesis is initiated by the transfer of the (R)-3-hydroxymyristate moiety from (R)-3-hydroxymyristoyl-acyl-carrier-protein (ACP) to the 3′-hydroxyl of UDP-GlcNAc to yield UDP-3-O-[(R)-3-hydroxymyristoyl]-N-acetyl-D-glucosamine (UDP-3-O-monoacyl-N-acetylglucosamine) (48). This reaction is catalyzed by the cytosolic enzyme UDP-N-acetylglucosamine acyltransferase (LpxA) (48,49). The IpxA gene has been cloned and sequenced, and it is located clockwise of dnaE at min 4 on the E. coli chromosome as part of a complex operon that includes three other genes involved in lipid A synthesis (lpxB,fabZ, and firA[lpxD]) as well as several additional open reading frames of unknown function (49–52). The deduced amino acid sequence of LpxA revealed it to be a 28 kDa protein comprised of 262 amino acids (50), and the enzyme has been purified to near homogeneity (53). Indeed, the x-ray crystal structure of the enzyme has been determined to 2.6 Å resolution, and an atomic model of LpxA based on the crystallographic data revealed the enzyme to be comprised of three identical subunits and a new kind of helical protein fold (54). The active site appears to be situated between the subunits. LpxA is the only acyltransferase for which a crystal structure is available.
Superoxide Dismutase, Mitochondrial Dysfunction, and Neurodegenerative Diseases
Published in Shamim I. Ahmad, Handbook of Mitochondrial Dysfunction, 2019
Jahaun Azadmanesh, Gloria E. O. Borgstahl
ALS prognosis correlates with the extent of IC-CuZnSOD aggregation into fibrils, much like amyloid-β fibrils in Alzheimer’s disease [70]. Fibrils form by IC-CuZnSOD misfolding and self-association into high-order, low energy structures. Examples are metal-deficient variants that destabilize the tertiary structure of the protein, and non-native interactions are energetically favored [116]. Other variants, affecting monomer structure and stability of the dimer interface, are thought to aggregate into fibrils as a result of lower thermostability and increased disorder [70,116]. Protein fold stability is further compromised through a lack of disulfide bond formation. Despite the cytosol being a strong reducing environment, IC-CuZnSOD harbors an oxidized disulfide bond that is buried near the dimer interface in WT. Variants that decrease dimer stability and expose the bond are more susceptible to disulfide reduction by glutathione/glutaredoxin systems [134]. A lack of Cu or Zn metallation also decreases dimer stability, meaning variants that compromise metal binding also expose the disulfide bond for reduction. The propensity for IC-CuZnSOD to form fibrils is related to factors compromising its stability, such as mutations and reduced disulfide bonds.
Artificial intelligence in early drug discovery enabling precision medicine
Published in Expert Opinion on Drug Discovery, 2021
Fabio Boniolo, Emilio Dorigatti, Alexander J. Ohnmacht, Dieter Saur, Benjamin Schubert, Michael P. Menden
The use of co-evolutionary sequence information has also led to new breakthroughs in protein structure prediction, as demonstrated in the two latest CASP competitions by AlphaFold [173]. This model uses predicted structural contacts from co-evolutionary models and refines them with a deep residual network that predicts the distribution of contact distances. These are interpreted as the statistical energy of the protein fold and directly minimized to yield highly accurate 3D protein structures. Since then, extensions have been made to remove the dependency on co-evolutionary analysis further [174], making it possible to predict 3D structures of artificial proteins and protein families starting from only a few sequences. Others have started to develop neural protein folding simulators that are fully trainable in an end-to-end-fashion and can directly map sequence to structure [175,176].
Emerging strategies to target the dysfunctional cohesin complex in cancer
Published in Expert Opinion on Therapeutic Targets, 2019
Konstantinos Mintzas, Michael Heuser
The cohesin complex consists of four core subunits that form a ring-shaped structure that encircles the DNA double strand (Figure 1). SMC1A (structural maintenance of chromosome protein 1A) and SMC3 are long polypeptides that form a V-shaped heterodimer[3]. Each protein folds back on itself to form a long anti-parallel coiled coil. On one end, the two ends of each protein come together and form the ATPase head, whereas on the other end, where the polypeptide folds, a dimerization domain joins the two proteins, thus creating a ‘hinge’. RAD21, a homolog of double-strand-break repair protein Rad21 from yeast, closes the circular structure by binding to the ATPase domains of SMC1A and SMC3 [2,4,5]. The fourth subunit of each cohesin complex is a member of the Stromal antigen proteins (STAG), either STAG1 or STAG2, which binds to the central region of RAD21 and stabilizes the complex. STAG2-containing complexes are more abundant in cells compared to STAG1-containing complexes and are located in the centromere regions of chromosomes. STAG1-containing complexes are responsible for cohesion of the chromatid arms in the non-centromeric and telomere regions[6]. Chromatin enters the ring through an entry gate that is formed by the transient dissociation of SMC1A and SMC3 in the hinge region and exits the structure through an exit gate at the meeting point of the SMC3 ATPase head and the N-terminus of RAD21 [7,8].
CALR-ETdb, the database of calreticulin variants diversity in essential thrombocythemia
Published in Platelets, 2022
Nora El Jahrani, Gabriel Cretin, Alexandre G. de Brevern
To improve the models, we added constraints on the positioning of the helices. The positions used were the positions with high confidence index. Type 1 DOPE scores decreased strongly, demonstrating better “globularity” in terms of the protein structure-sequence relationship. However, the best DOPE score had an unexpected additional helix, so another model for type 1 with a very close DOPE score was selected. For the type 2 models, the best DOPE score was still associated with a misplaced helix, so another model with a close DOPE score was chosen, but the model was therefore structurally further away. The two models selected consequently exhibited protein folds with helices corresponding to the expected data (see Figure 4a and b).