Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “MUTATIONS”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2021

Association of somatic mutations in BRCA2 BRC domain with chemotherapy sensitivity and survival in high grade serous ovarian cancer

Highlights: • BRCA2 BRC domain-associated mutation confers higher platinum sensitivity of HGSOC. • Somatic BRCA2 BRC domain mutation impairs BRCA2 HR in HGSOC. • BRCA2 BRC domain mutation at sites is crucial for HR repair of DNA DSB induced by platinums. • Somatic BRCA2 BRC domain missense mutations improve outcomes in patients with HGSOC. Mutations at sites crucial for the interaction between RAD51 and BRC domains impair the ability of BRCA2 homologous recombination. We aimed to clarify whether BRCA2 BRC domain-associated mutation correlates with sensibility of platinum-based chemotherapy and survival in high-grade serous ovarian cancer (HGSOC).

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Isolation and characterization of novel mutations in the pSC101 origin that increase copy number

pSC101 is a narrow host range, low-copy plasmid commonly used for genetically manipulating Escherichia coli. As a byproduct of a genetic screen for a more sensitive lactam biosensor, we identified multiple novel mutations that increase the copy number of plasmids with the pSC101 origin. All mutations identified in this study occurred on plasmids which also contained at least one mutation localized to the RepA protein encoded within the origin. Homology modelling predicts that many of these mutations occur within the dimerization interface of RepA. Mutant RepA resulted in plasmid copy numbers between ~31 and ~113 copies/cell, relative to ~5 copies/cell in wild-type pSC101 plasmids. Combining the mutations that were predicted to disrupt multiple contacts on the dimerization interface resulted in copy numbers of ~500 copies/cell, while also attenuating growth in host strains. Fluorescent protein production expressed from an arabinose-inducible promoter on mutant origin derived plasmids did correlate with copy number. Plasmids harboring RepA with one of two mutations, E83K and N99D, resulted in fluorescent protein production similar to that from p15a- (~20 copies/cell) and ColE1- (~31 copies/cell) based plasmids, respectively. The mutant copy number variants retained compatibility with p15a, pBBR, and ColE1 origins of replication. Thus, these pSC101 variants may be useful in future metabolic engineering efforts that require medium or high-copy vectors compatible with p15a- and ColE1-based plasmids.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2019

Functional divergence caused by mutations in an energetic hotspot in ERK2

The most frequent extracellular signal-regulated kinase 2 (ERK2) mutation occurring in cancers is E322K (E-K). ERK2 E-K reverses a buried charge in the ERK2 common docking (CD) site, a region that binds activators, inhibitors, and substrates. Little is known about the cellular consequences associated with this mutation, other than apparent increases in tumor resistance to pathway inhibitors. ERK2 E-K, like the mutation of the preceding aspartate (ERK2 D321N [D-N]) known as the sevenmaker mutation, causes increased activity in cells and evades inactivation by dual-specificity phosphatases. As opposed to findings in cancer cells, in developmental assays in Drosophila , only ERK2 D-N displays a significant gain of function, revealing mutation-specific phenotypes. The crystal structure of ERK2 D-N is indistinguishable from that of wild-type protein, yet this mutant displays increased thermal stability. In contrast, the crystal structure of ERK2 E-K reveals profound structural changes, including disorder in the CD site and exposure of the activation loop phosphorylation sites, which likely account for the decreased thermal stability of the protein. These contiguous mutations in the CD site of ERK2 are both required for docking interactions but lead to unpredictably different functional outcomes. Overall, our results suggest that the CD site is in an energetically strained configuration, and this helps drive conformational changes at distal sites on ERK2 during docking interactions.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2018

Predicting the Effect of Single and Multiple Mutations on Protein Structural Stability

Predicting how a point mutation alters a protein’s stability can guide pharmaceutical drug design initiatives which aim to counter the effects of serious diseases. Conducting mutagenesis studies in physical proteins can give insights about the effects of amino acid substitutions, but such wet-lab work is prohibitive due to the time as well as financial resources needed to assess the effect of even a single amino acid substitution. Computational methods for predicting the effects of a mutation on a protein structure can complement wet-lab work, and varying approaches are available with promising accuracy rates. In this work we compare and assess the utility of several machine learning methods and their ability to predict the effects of single and double mutations. We in silico generate mutant protein structures, and compute several rigidity metrics for each of them. We use these as features for our Support Vector Regression (SVR), Random Forest (RF), and Deep Neural Network (DNN) methods. We validate the predictions of our in silico mutations against experimental ΔΔG stability data, and attain Pearson Correlation values upwards of 0.71 for single mutations, and 0.81 for double mutations. We perform ablation studies to assess which features contribute most to a model’s success, and also introduce a voting scheme to synthesize a single prediction from the individual predictions of the three models.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
NASA NTRS2001

A comparison of mutations induced by accelerated iron particles versus those induced by low earth orbit space radiation in the FEM-3 gene of Caenorhabditis elegans

The fem-3 gene of Caenorhabditis elegans was employed to determine the mutation frequency as well as the nature of mutations induced by low earth orbit space radiation ambient to Space Shuttle flight STS-76. Recovered mutations were compared to those induced by accelerated iron ions generated by the AGS synchrotron accelerator at Brookhaven National Laboratory. For logistical reasons, dauer larvae were prepared at TCU, transported to either Kennedy Space Center or Brookhaven National Laboratory, flown in space or irradiated, returned to TCU and screened for mutants. A total of 25 fem-3 mutants were recovered after the shuttle flight and yielded a mutation frequency of 2.1x10(-5), roughly 3.3-fold higher than the spontaneous rate of 6.3x10(-6). Four of the mutations were homozygous inviable, suggesting that they were large deletions encompassing fem-3 as well as neighboring, essential genes. Southern blot analyses revealed that one of the 25 contained a polymorphism in fem-3, further evidence that space radiation can induce deletions. While no polymorphisms were detected among the iron ion-induced mutations, three of the 15 mutants were homozygous inviable, which is in keeping with previous observations that high LET iron particles generate deficiencies. These data provide evidence, albeit indirect, that an important mutagenic component of ambient space radiation is high LET charged particles such as iron ions.

short duration↗
NASA NTRS1993

Diploid yeast cells yield homozygous spontaneous mutations

A leucine-requiring hybrid of Saccharomyces cerevisiae, homoallelic at the LEU1 locus (leu1-12/leu1-12) and heterozygous for three chromosome-VII genetic markers distal to the LEU1 locus, was employed to inquire: (1) whether spontaneous gene mutation and mitotic segregation of heterozygous markers occur in positive nonrandom association and (2) whether homozygous LEU1/LEU1 mutant diploids are generated. The results demonstrate that gene mutation of leu1-12 to LEU1 and mitotic segregation of heterozygous chromosome-VII markers occur in strong positive nonrandom association, suggesting that the stimulatory DNA lesion is both mutagenic and recombinogenic. In addition, genetic analysis of diploid Leu+ revertants revealed that approximately 3% of mutations of leu1-12 to LEU1 result in LEU1/LEU1 homozygotes. Red-white sectored Leu+ colonies exhibit genotypes that implicate post-replicational chromatid breakage and exchange near the site of leu1-12 reversion, chromosome loss, and subsequent restitution of diploidy, in the sequence of events leading to mutational homozygosis. By analogy, diploid cell populations can yield variants homozygous for novel recessive gene mutations at biologically significant rates. Mutational homozygosis may be relevant to both carcinogenesis and the evolution of asexual diploid organisms.

Non-NASA Center↗
DOE OSTI2018

Myoclonus epilepsy, retinitis pigmentosa, leukoencephalopathy and cerebral calcifications associated with a novel m.5513G>A mutation in the MT-TW gene

Highlights: • Sequencing the mtDNA from a woman with mitochondrial encephalomyopathy. • Patient showing a novel heteroplasmic mutation at nucleotide 5513 in the RNA{sup Trp}. • 17° mutation in MT-TW gene expanding the known causes of mitochondrial disease. We sequenced the mitochondrial genome from a 40-year-old woman with myoclonus epilepsy, retinitis pigmentosa, leukoencephalopathy and cerebral calcifications. Histological and biochemical features of mitochondrial respiratory chain dysfunction were present. Direct sequencing showed a novel heteroplasmic mutation at nucleotide 5513 in the MT-TW gene that encodes tRNA{sup Trp}. Restriction Fragment Length Polymorphism analysis confirmed that about 80% of muscle mtDNA harboured the mutation while it was present in minor percentages in mtDNA from other tissues. The mutation is predicted to disrupt a highly conserved base pair within the aminoacyl acceptor stem of the tRNA. This is the 17° mutation in MT-TW gene and expands the known causes of late-onset mitochondrial diseases.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Fam83h mutation inhibits the mineralization in ameloblasts by activating Wnt/β-catenin signaling pathway

Highlights: • Fam83h mutation altered the expression and subcellular localization of Fam83 h, CK1α and β-catenin. • Fam83h mutation activated the Wnt/β-catenin signaling pathway and inhibited the mineralization in LS8 cells. • The inhibitor of Wnt/β-catenin signaling pathway rescued the inhibited mineralization in LS8 cells with mutant Fam83h. • Fam83h mutation could inhibit the mineralization in ameloblasts by activating Wnt/β-catenin signaling pathway. FAM83H was identified as the major causative gene for autosomal dominant hypocalcified amelogenesis imperfect (ADHCAI). The pathogenic mechanism of FAM83H in ADHCAI remains elusive. The present study aims to investigate the effect of Fam83h mutation on the mineralization of mouse ameloblast cell line LS8 and to explore the possible pathogenesis of ADHCAI. Lentivirus package was performed for the plasmids with mouse Fam83h mutant cDNA (c.1186C > T, M3) and empty vector (Control) and transfected into LS8, which were divided into M3-FLAG and Control groups. Immunoprecipitation, western-blot and immunofluorescence were performed to detect the expression and subcellular localization of Fam83 h, CK1α and β-catenin. ALP activity, ALP staining, expression of the mineralization factors were detected in two groups during mineralization induction. Expression of the mineralization factors was also detected in M3-FLAG and LS8 exposing to pyrvinium pamoate. Compared with the Control, the Fam83h mutation altered the expression and localization of Fam83 h, CK1α and β-catenin in LS8, inhibited the mineralization and down-regulated the expression of mineralization factors in M3-FLAG. Pyrvinium pamoate, an inhibitor of the Wnt/β-catenin signaling pathway, up-regulated expression of mineralization factors in LS8 and rescued the inhibited mineralization in M3-FLAG. The results indicated that the Fam83h mutation could inhibit the mineralization in ameloblasts by activating Wnt/β-catenin signaling pathway.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Early emergence of de novo EGFR T790M gatekeeper mutations during erlotinib treatment in PC9 non-small cell lung cancer cells

The emergence of the T790M gatekeeper mutation in the Epidermal Growth Factor Receptor (EGFR) gene is an important mechanism that can lead to the acquired resistance to EGFR-targeted tyrosine kinase inhibitors such as erlotinib or gefitinib. These drugs have been used in treating a subset of non-small cell lung cancer (NSCLC) patients harboring EGFR activating mutations. Here we investigated the paths leading to the acquisition of the T790M mutation by establishing an erlotinib resistant PC9 cell model harboring ectopically introduced EGFR cDNA. We detected the emergence of T790M mutation within the EGFR cDNA in a subset of erlotinib resistant PC9 cell models through Sanger sequencing and droplet digital PCR-based methods, demonstrating that T790M mutation can emerge via de novo events following treatment with erlotinib. In addition, we show that the de novo T790M bearing erlotinib resistant PC9 cells are sensitive to the 3rd generation EGFR-targeted drug, WZ4002. Furthermore, GFP-based competition cell proliferation assays reveal that PC9 cells ectopically expressing EGFR mutant become more rapidly resistant to erlotinib than parental PC9 cells through the acquisition of the T790M mutation. Taken together, we believe that our findings expand upon the previous notion of evolutionary paths of T790M development, providing an important clue to designing a therapeutic strategy to overcome drug resistance.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

TTF-1/Nkx2.1 functional connection with mutated EGFR relies on LRIG1 and β-catenin pathways in lung cancer cells

Highlights: • In lung cancer TTF-1/Nkx2.1 expression correlates with the presence of EGFR mutations. • TTF-1/Nkx2.1 down-regulating LRIG1 gene expression hinders EGFR degradation. • β-catenin pathway activates TTF-1/Nkx2.1 gene expression. • Mutated EGFR sustains TTF-1/Nkx2.1 expression via β-catenin pathway. In non-small lung cancer, the expression of the transcription factor TTF-1/Nkx2.1 correlates with the presence of EGFR mutations, therefore TTF-1/Nkx2.1 expression is used to optimize an EGFR testing strategy and to guide clinical treatment. We investigate the molecular mechanisms underlying the functional connection between EGFR and TTF-1/Nkx2.1 gene expression in lung adenocarcinoma. Using the H1975 cell line as a non-small cell lung cancer model system and short hairpin RNA, we have selected clones with TTF-1/Nkx2.1 silenced expression. We have found that Leucine-rich immunoglobulin repeats-1 (LRIG1) gene is a direct target of TTF-1/Nkx2.1 and the transcription factor binding to the LRIG1 genomic sequence inhibits its gene expression. In TTF-1/Nkx2.1 depleted clones, we have found high levels of LRIG1 and decreased presence of EGFR protein. Furthermore, in TTF-1/Nkx2.1 depleted clones we detected a reduced β-catenin level and we provide experimental evidence indicating that TTF-1/Nkx2.1 gene expression is regulated by β-catenin. Published studies indicate that LRIG1 triggers EGFR degradation and that mutated EGFR induces β-catenin activity. Hence, with the present study we show that mutated EGFR, enhancing β-catenin, stimulates TTF-1/Nkx2.1 gene expression and, at the same time, TTF-1/Nkx2.1, down-regulating LRIG1, sustains EGFR pathway. Therefore, LRIG1 and β-catenin mediate the functional connection between TTF-1/Nkx2.1 and mutated EGFR.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2018

Structural reorganization of SHP2 by oncogenic mutations and implications for oncoprotein resistance to allosteric inhibition

Activating mutations in PTPN11, encoding the cytosolic protein tyrosine phosphatase SHP2, result in developmental disorders and act as oncogenic drivers in patients with hematologic cancers. The allosteric inhibitor SHP099 stabilizes the wild-type SHP2 enzyme in an autoinhibited conformation that is itself destabilized by oncogenic mutations. Here, we report the impact of the highly activated and most frequently observed mutation, E76K, on the structure of SHP2, and investigate the effect of E76K and other oncogenic mutations on allosteric inhibition by SHP099. SHP2 E76K adopts an open conformation but can be restored to the closed, autoinhibited conformation, near-identical to the unoccupied wild-type enzyme, when complexed with SHP099. SHP099 inhibitory activity against oncogenic SHP2 variants in vitro and in cells scales inversely with the activating strength of the mutation, indicating that either oncoselective or vastly more potent inhibitors will be necessary to suppress oncogenic signaling by the most strongly activating SHP2 mutations in cancer.

Chemical biology↗
DOE OSTI2020

Structural basis for impairment of DNA methylation by the DNMT3A R882H mutation

DNA methyltransferase DNMT3A is essential for establishment of mammalian DNA methylation during development. The R882H DNMT3A is a hotspot mutation in acute myeloid leukemia (AML) causing aberrant DNA methylation. However, how this mutation affects the structure and function of DNMT3A remains unclear. Here we report structural characterization of wild-type and R882H-mutated DNMT3A in complex with DNA substrates with different sequence contexts. A loop from the target recognition domain (TRD loop) recognizes the CpG dinucleotides in a +1 flanking site-dependent manner. The R882H mutation reduces the DNA binding at the homodimeric interface, as well as the molecular link between the homodimeric interface and TRD loop, leading to enhanced dynamics of TRD loop. Consistently, in vitro methylation analyses indicate that the R882H mutation compromises the enzymatic activity, CpG specificity and flanking sequence preference of DNMT3A. Together, this study uncovers multiple defects of DNMT3A caused by the R882H mutation in AML.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Deep mutational scanning reveals the molecular determinants of RNA polymerase-mediated adaptation and tradeoffs

RNA polymerase (RNAP) is emblematic of complex biological systems that control multiple traits involving trade-offs such as growth versus maintenance. Laboratory evolution has revealed that mutations in RNAP subunits, including RpoB, are frequently selected. However, we lack a systems view of how mutations alter the RNAP molecular functions to promote adaptation. We, therefore, measured the fitness of thousands of mutations within a region of rpoB under multiple conditions and genetic backgrounds, to find that adaptive mutations cluster in two modules. Mutations in one module favor growth over maintenance through a partial loss of an interaction associated with faster elongation. Mutations in the other favor maintenance over growth through a destabilized RNAP-DNA complex. The two molecular handles capture the versatile RNAP-mediated adaptations. Combining both interaction losses simultaneously improved maintenance and growth, challenging the idea that growth-maintenance tradeoff resorts only from limited resources, and revealing how compensatory evolution operates within RNAP.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Structure-guided functional suppression of AML-associated DNMT3A hotspot mutations

DNA methyltransferases DNMT3A- and DNMT3B-mediated DNA methylation critically regulate epigenomic and transcriptomic patterning during development. The hotspot DNMT3A mutations at the site of Arg822 (R882) promote polymerization, leading to aberrant DNA methylation that may contribute to the pathogenesis of acute myeloid leukemia (AML). However, the molecular basis underlying the mutation-induced functional misregulation of DNMT3A remains unclear. Here, we report the crystal structures of the DNMT3A methyltransferase domain, revealing a molecular basis for its oligomerization behavior distinct to DNMT3B, and the enhanced intermolecular contacts caused by the R882H or R882C mutation. Our biochemical, cellular, and genomic DNA methylation analyses demonstrate that introducing the DNMT3B-converting mutations inhibits the R882H-/R882C-triggered DNMT3A polymerization and enhances substrate access, thereby eliminating the dominant-negative effect of the DNMT3A R882 mutations in cells. Together, this study provides mechanistic insights into DNMT3A R882 mutations-triggered aberrant oligomerization and DNA hypomethylation in AML, with important implications in cancer therapy.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2020

Human XPG nuclease structure, assembly, and activities with insights for neurodegeneration and cancer from pathogenic mutations

Xeroderma pigmentosum group G (XPG) protein is both a functional partner in multiple DNA damage responses (DDR) and a pathway coordinator and structure-specific endonuclease in nucleotide excision repair (NER). Different mutations in the XPG gene ERCC5 lead to either of two distinct human diseases: Cancer-prone xeroderma pigmentosum (XP-G) or the fatal neurodevelopmental disorder Cockayne syndrome (XP-G/CS). To address the enigmatic structural mechanism for these differing disease phenotypes and for XPG’s role in multiple DDRs, here we determined the crystal structure of human XPG catalytic domain (XPGcat), revealing XPG-specific features for its activities and regulation. Furthermore, XPG DNA binding elements conserved with FEN1 superfamily members enable insights on DNA interactions. Notably, all but one of the known pathogenic point mutations map to XPGcat, and both XP-G and XP-G/CS mutations destabilize XPG and reduce its cellular protein levels. Mapping the distinct mutation classes provides structure-based predictions for disease phenotypes: Residues mutated in XP-G are positioned to reduce local stability and NER activity, whereas residues mutated in XP-G/CS have implied long-range structural defects that would likely disrupt stability of the whole protein, and thus interfere with its functional interactions. Combined data from crystallography, biochemistry, small angle X-ray scattering, and electron microscopy unveil an XPG homodimer that binds, unstacks, and sculpts duplex DNA at internal unpaired regions (bubbles) into strongly bent structures, and suggest how XPG complexes may bind both NER bubble junctions and replication forks. Collective results support XPG scaffolding and DNA sculpting functions in multiple DDR processes to maintain genome stability.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Structural basis for severe pain caused by mutations in the S4-S5 linkers of voltage-gated sodium channel Na V 1.7

Gain-of-function mutations in voltage-gated sodium channel Na V 1.7 cause severe inherited pain syndromes, including inherited erythromelalgia (IEM). The structural basis of these disease mutations, however, remains elusive. Here, we focused on three mutations that all substitute threonine residues in the alpha-helical S4-S5 intracellular linker that connects the voltage sensor to the pore: Na V 1.7/I234T, Na V 1.7/I848T, and Na V 1.7/S241T in order of their positions in the amino acid sequence within the S4-S5 linkers. Introduction of these IEM mutations into the ancestral bacterial sodium channel Na V Ab recapitulated the pathogenic gain-of-function of these mutants by inducing a negative shift in the voltage dependence of activation and slowing the kinetics of inactivation. Remarkably, our structural analysis reveals a common mechanism of action among the three mutations, in which the mutant threonine residues create new hydrogen bonds between the S4-S5 linker and the pore-lining S5 or S6 segment in the pore module. Because the S4-S5 linkers couple voltage sensor movements to pore opening, these newly formed hydrogen bonds would stabilize the activated state substantially and thereby promote the 8 to 18 mV negative shift in the voltage dependence of activation that is characteristic of the Na V 1.7 IEM mutants. Our results provide key structural insights into how IEM mutations in the S4-S5 linkers may cause hyperexcitability of Na V 1.7 and lead to severe pain in this debilitating disease.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2017

Structure of human POFUT1, its requirement in ligand-independent oncogenic Notch signaling, and functional effects of Dowling-Degos mutations

Protein O-fucosyltransferase-1 (POFUT1), which transfers fucose residues to acceptor sites on serine and threonine residues of epidermal growth factor-like repeats of recipient proteins, is essential for Notch signal transduction in mammals. Here, we examine the consequences of POFUT1 loss on the oncogenic signaling associated with certain leukemia-associated mutations of human Notch1, report the structures of human POFUT1 in free and GDP-fucose bound states, and assess the effects of Dowling-Degos mutations on human POFUT1 function. CRISPR-mediated knockout of POFUT1 in U2OS cells suppresses both normal Notch1 signaling, and the ligand-independent signaling associated with leukemogenic mutations of Notch1. Normal and oncogenic signaling are rescued by wild-type POFUT1 but rescue is impaired by an active-site R240A mutation. The overall structure of the human enzyme closely resembles that of the Caenorhabditis elegans protein, with an overall backbone RMSD of 0.93 Å, despite primary sequence identity of only 39% in the mature protein. GDP-fucose binding to the human enzyme induces limited backbone conformational movement, though the side chains of R43 and D244 reorient to make direct contact with the fucose moiety in the complex. The reported Dowling-Degos mutations of POFUT1, except for M262T, fail to rescue Notch1 signaling efficiently in the CRISPR-engineered POFUT1 -/- background. Together, these studies identify POFUT1 as a potential target for cancers driven by Notch1 mutations and provide a structural roadmap for its inhibition.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2022

Automating Genetic Algorithm Mutations for Molecules Using a Masked Language Model

Inspired by the evolution of biological systems, genetic algorithms have been applied to generate solutions for optimization problems in a variety of scientific and engineering disciplines. For a given problem, a suitable genome representation must be defined along with a mutation operator to generate subsequent generations. Unlike natural systems which display a variety of complex rearrangements (e.g. mobile genetic elements), mutation for genetic algorithms commonly utilizes only random point-wise changes. Furthermore, generalizing beyond point-wise mutations poses a key difficulty as useful genome rearrangements depend on the representation and problem domain. To move beyond the limitations of manually defined point-wise changes, here we propose the use of techniques from masked language models to automatically generate mutations. As a first step, common subsequences within a given population are used to generate a vocabulary. The vocabulary is then used to tokenize each genome. A masked language model is trained on the tokenized data in order to generate possible rearrangements (i.e. mutations). In order to illustrate the proposed strategy, we use string representations of molecules and use a genetic algorithm to optimize for drug-likeness and synthesizability. Finally, our results show that moving beyond random point-wise mutations accelerates genetic algorithm optimization.

59 BASIC BIOLOGICAL SCIENCES↗