Biology subjects

Ma, X.

Publications and source records attributed to Ma, X..

At least 253 records · Page 14Linked to original sources

bioRxiv2023

Normative modeling of brain morphometry in Clinical High-Risk for Psychosis

ImportanceThe lack of robust neuroanatomical markers of psychosis risk has been traditionally attributed to heterogeneity. A complementary hypothesis is that variation in neuroanatomical measures in the majority of individuals at psychosis risk may be nested within the range observed in healthy individuals. ObjectiveTo quantify deviations from the normative range of neuroanatomical variation in individuals at clinical high-risk for psychosis (CHR-P) and evaluate their overlap with healthy variation and their association with positive symptoms, cognition, and conversion to a psychotic disorder. Design, Setting, and ParticipantsClinical, IQ and FreeSurfer-derived regional measures of cortical thickness (CT), cortical surface area (SA), and subcortical volume (SV) from 1,340 CHR-P individuals [47.09% female; mean age: 20.75 (4.74) years] and 1,237 healthy individuals [44.70% female; mean age: 22.32 (4.95) years] from 29 international sites participating in the ENIGMA Clinical High Risk for Psychosis Working Group. Main Outcomes and MeasuresFor each regional morphometric measure, z-scores were computed that index the degree of deviation from the normative means of that measure in a healthy reference population (N=37,407). Average deviation scores (ADS) for CT, SA, SV, and globally across all measures (G) were generated by averaging the respective regional z-scores. Regression analyses were used to quantify the association of deviation scores with clinical severity and cognition and two-proportion z-tests to identify case-control differences in the proportion of individuals with infranormal (z<-1.96) or supranormal (z>1.96) scores. ResultsCHR-P and healthy individuals overlapped in the distributions of the observed values, regional z-scores, and all ADS vales. The proportion of CHR-P individuals with infranormal or supranormal values in any metric was low (<12%) and similar to that of healthy individuals. CHR-P individuals who converted to psychosis compared to those who did not convert had a higher percentage of infranormal values in temporal regions (5-7% vs 0.9-1.4%). In the CHR-P group, only the ADSSA showed significant but weak associations (|{beta}|<0.09; PFDR<0.05) with positive symptoms and IQ. Conclusions and RelevanceThe study findings challenge the usefulness of macroscale neuromorphometric measures as diagnostic biomarkers of psychosis risk and suggest that such measures do not provide an adequate explanation for psychosis risk. Key pointsO_ST_ABSQuestionC_ST_ABSIs the risk of psychosis associated with brain morphometric changes that deviate significantly from healthy variation? FindingsIn this study of 1340 individuals high-risk for psychosis (CHR-P) and 1237 healthy participants, individual-level variation in macroscale neuromorphometric measures of the CHR-P group was largely nested within healthy variation and was not associated with the severity of positive psychotic symptoms or conversion to a psychotic disorder. MeaningThe findings suggest the macroscale neuromorphometric measures have limited utility as diagnostic biomarkers of psychosis risk.

neuroscience↗

bioRxiv2023

Engineering Escherichia coli to produce aromatic chemicals from ethylene glycol

Microbial overproduction of aromatic chemicals has gained considerable industrial interest and various metabolic engineering approaches have been employed in recent years to address the associated challenges. So far, most studies have used sugars (mostly glucose) or glycerol as the primary carbon source. In this study, we used ethylene glycol (EG) as the main carbon substrate. EG could be obtained from the degradation of plastic and cellulosic wastes. As a proof of concept, Escherichia coli was engineered to transform EG into L-tyrosine, a valuable aromatic amino acid. Under the best fermentation condition, the strain produced 2 g/L L-tyrosine from 10 g/L EG at approximately 50% of the theoretical yield, outperforming glucose (the most common sugar feedstock) in the same experimental conditions. To prove the concept that EG can be converted into different aromatic chemicals, E. coli was further engineered with a similar approach to synthesize other valuable aromatic chemicals, L-phenylalanine and p-coumaric acid. Finally, waste polyethylene terephthalate (PET) bottles were degraded using acid hydrolysis and the resulting monomer EG was transformed into L-tyrosine using the engineered E. coli, yielding a comparable titer to that obtained using commercial EG. The strains developed in this study should be valuable to the community for producing valuable aromatics from EG.

bioengineering↗

bioRxiv2023

Combined genome-scale fitness and paralog synthetic lethality screens with just 44k clones: the IN4MER CRISPR/Cas12a multiplex knockout platform

Genetic interactions mediate the emergence of phenotype from genotype, but initial technologies for combinatorial genetic perturbation in mammalian cells suffer from inefficiency and are challenging to scale. Recent focus on paralog synthetic lethality in cancer cells offers an opportunity to evaluate different approaches and improve on the state of the art. Here we report a meta-analysis of CRISPR genetic interactions screens, identifying a candidate set of background-independent paralog synthetic lethals, and find that the Cas12a platform provides superior sensitivity and assay replicability. We demonstrate that Cas12a can independently target up to four genes from a single guide array, and we build on this knowledge by constructing a genome-scale library that expresses arrays of four guides per clone, a platform we call in4mer. Our genome-scale human library, with only 49k clones, is substantially smaller than a typical CRISPR/Cas9 monogenic library while also targeting more than four thousand paralog pairs, triples, and quads. Proof of concept screens in four cell lines demonstrate discrimination of core and context-dependent essential genes similar to that of state-of-the-art CRISPR/Cas9 libraries, as well as detection of synthetic lethal and masking/buffering genetic interactions between paralogs of various family sizes, a capability not offered by any extant library. Importantly, the in4mer platform offers a fivefold reduction in the number of clones required to assay genetic interactions, dramatically improving the cost and effort required for these studies.

systems biology↗

bioRxiv2022

Cataloguing and profiling of the gut virome in Chinese populations uncover extensive viral signatures across common diseases

The gut viral community has been linked to human physiology and health, but our knowledge of its genetic and functional contents and disease dependence is far from complete. Here, we collected 11,327 bulk or viral metagenomes from fecal samples from large-scale Chinese populations to establish a Chinese gut virus catalogue (cnGVC) comprising 67,096 nonredundant viral genomes. This catalogue included [~]70% of novel viruses that are not represented in existing gut viral databases, and allowed us to characterize the functional diversity and specificity of the gut virome. Using cnGVC, we 1) profiled the gut virome in large-scale populations and evaluated their sex- and age-related variations, 2) investigated the diversity and compositional patterns of the gut virome across common diseases by analyzing 6,314 bulk metagenomes spanning 28 disease or unhealthy statuses, and 3) identified a large number of universal viral signatures of diseases and validated their predictive ability for health status. Overall, our resources and results would contribute to the grand effort of expanding the knowledge of the human gut virome and addressing a full picture of the associations between viruses and common diseases.

microbiology↗

bioRxiv2022

ZBTB20 is Essential for Cochlear Maturation and Hearing in Mice

The mammalian cochlear epithelium undergoes substantial remodeling and maturation before the onset of hearing. However, very little is known about the transcriptional network governing cochlear late-stage maturation and particularly the differentiation of its lateral non-sensory region. Here we establish ZBTB20 as an essential transcription factor required for cochlear terminal differentiation and maturation and hearing. ZBTB20 is abundantly expressed in the developing and mature cochlear non-sensory epithelial cells, with transient expression in immature hair cells and spiral ganglion neurons. Otocyst-specific deletion of Zbtb20 causes profound deafness with reduced endolymph potential in mice. The subtypes of cochlear epithelial cells are normally generated but their postnatal development is arrested in the absence of ZBTB20, as manifested by an immature appearance of the organ of Corti, malformation of tectorial membrane, a flattened spiral prominence, and a lack of identifiable Boettcher cells. Furthermore, these defects are related with a failure in the terminal differentiation of the non-sensory epithelium covering the outer border Claudius cells, outer sulcus root cells and spiral prominence epithelial cells. Transcriptome analysis shows ZBTB20 regulates genes coding for tectorial membrane proteins in the greater epithelial ridge, and those preferentially expressed in root cells and spiral prominence epithelium. Our results point to ZBTB20 as an essential regulator for postnatal cochlear maturation and particularly for the terminal differentiation of cochlear lateral non-sensory domain.

developmental biology↗

bioRxiv2022

COVID-19db linkage maps of cell surface proteins and transcription factors in immune cells

The highly contagious SARS-CoV-2 and its associated disease (COVID-19) are a threat to global public health and economies. To develop effective treatments for COVID-19, we must understand the host cell types, cell states and regulators associated with infection and pathogenesis such as dysregulated transcription factors (TFs) and surface proteins, including signalling receptors. To link cell surface proteins with TFs, we recently developed SPaRTAN (Single-cell Proteomic and RNA-based Transcription factor Activity Network) by integrating parallel single-cell proteomic and transcriptomic data based on Cellular Indexing of Transcriptomes and Epitopes by sequencing (CITE-seq) and gene cis-regulatory information. We apply SPaRTAN to CITE-seq datasets from patients with varying degrees of COVID-19 severity and healthy controls to identify the associations between surface proteins and TFs in host immune cells. Here, we present COVID-19db of Immune Cell States (https://covid19db.streamlit.app/), a web server containing cell surface protein expression, SPaRTAN-inferred TF activities, and their associations with major host immune cell types. The data include four high-quality COVID-19 CITE-seq datasets with a toolset for userfriendly data analysis and visualization. We provide interactive surface protein and TF visualizations across major immune cell types for each dataset, allowing comparison between various patient severity groups for the discovery of potential therapeutic targets and diagnostic biomarkers.

bioinformatics↗

bioRxiv2022

A tRNA Modification-based strategy for Identifying amiNo acid Overproducers (AMINO)

Amino acids have a multi-billion-dollar market with rising demand, prompting the development of high-performance microbial factories. However, a general screening strategy applicable to all proteinogenic and non-proteinogenic amino acids is still lacking. Modification of the critical structure of tRNA could decrease the aminoacylation level of tRNA catalyzed by aminoacyl-tRNA synthetases. Involved in a two-substrate sequential reaction, amino acids with increased concentration could elevate the reduced aminoacylation rate caused by specific tRNA modification. Here, we developed a selection system for overproducers of specific amino acids using corresponding engineered tRNAs and reporter genes. As a proof-of-concept, overproducers of five amino acids such as L-tryptophan were screened out by growth-based and/or fluorescence-activated cell sorting (FACS)-based screening from random mutation libraries of Escherichia coli and Corynebacterium glutamicum, respectively. This study provided a universal strategy that could be applied to screen overproducers of proteinogenic and non-proteinogenic amino acids in amber-stop-codon-recoded or non-recoded hosts.

synthetic biology↗

bioRxiv2022

Monkey-to-human transfer of brain computer interface decoders

Intracortical brain-computer interfaces (iBCIs) enable paralyzed persons to generate movement, but current methods require large amounts of both neural and movement-related data to be collected from the iBCI user for supervised decoder training. We hypothesized that the low-dimensional latent neural representations of motor behavior, known to be preserved across time, might also be preserved across individuals, and allow us to circumvent this problem. We trained a decoder to predict the electromyographic (EMG) activity for a "source" monkey from the latent signals of motor cortex. We then used Canonical Correlation Analysis to align the latent signals of a "target" monkey to those of the source. These decoders were as accurate across monkeys as they were across sessions for a given monkey. Remarkably, the same process with latent signals from a human participant with tetraplegia was within 90% of the with-monkey decoding across session accuracy. Our findings suggest that consistent representations of motor activity exist across animals and even species. Discovering this common representation is a crucial first step in designing iBCI decoders that perform well without large amounts of data and supervised subject-specific tuning.

neuroscience↗

bioRxiv2022

Evaluation of antibody kinetics and durability in health subjects vaccinated with inactivated COVID-19 vaccine (CoronaVac): A cross-sectional and cohort study in Zhejiang, China

BackgroundAlthough inactivated COVID-19 vaccines are proven to be safe and effective in the general population, the dynamic response and duration of antibodies after vaccination in the real world should be further assessed. MethodsWe enrolled 1067 volunteers who had been vaccinated with one or two doses of CoronaVac in Zhejiang Province, China. Another 90 healthy adults without previous vaccinations were recruited and vaccinated with three doses of CoronaVac, 28 days and 6 months apart. Serum samples were collected from multiple timepoints and analyzed for specific IgM/IgG and neutralizing antibodies (NAbs) for immunogenicity evaluation. Antibody responses to the Delta and Omicron variants were measured by pseudovirus-based neutralization tests. ResultsOur results revealed that binding antibody IgM peaked 14-28 days after one dose of CoronaVac, while IgG and NAbs peaked approximately 1 month after the second dose then declined slightly over time. Antibody responses had waned by month 6 after vaccination and became undetectable in the majority of individuals at 12 months. Levels of NAbs to live SARS-CoV-2 were correlated with anti-SARS-CoV-2 IgG and NAbs to pseudovirus, but not IgM. Homologous booster around 6 months after primary vaccination activated anamnestic immunity and raised NAbs 25.5-fold. The NAb inhibition rate subsequently rose to 36.0% for Delta (p=0.03) and 4.3% for Omicron (p=0.004), and the response rate for Omicron rose from 7.9% (7/89) to 17.8% (16/90). ConclusionsTwo doses of CoronaVac vaccine resulted in limited protection over a short duration. The homologous booster slightly increased antibody responses to the Delta and Omicron variants; therefore, the optimization of booster procedures is vital. FundingKey Research and Development Program of Zhejiang Province; Key Program of Health Commission of Zhejiang Province/ Science Foundation of National Health Commission; Major Program of Zhejiang Municipal Natural Science Foundation.

immunology↗

bioRxiv2022

Structure of the human UBR5 E3 ubiquitin ligase

The human UBR5 (also known as EDD) is a single polypeptide chain HECT-type E3 ubiquitin ligase essential for embryonic development in mammals. Although widely expressed, UBR5 is markedly amplified and overexpressed in breast, ovarian, prostate, gastric and pancreatic cancers. Dysregulated UBR5 functions like an oncoprotein to promote cancer growth and metastasis, making UBR5 a potential target for therapeutics. Unexpectedly, we found that human UBR5 assembles a dimer and a tetramer in solution. We determined the dimer structure at 2.8 [A] and the tetramer structure at 3.5 [A] average resolution. UBR5 is a crescent shaped molecule with a seven-bladed {beta}-propeller and two small {beta}-barrel domains (SBB1/2) at the N-terminal region, a catalytic HECT domain at the C-terminus, and an extended helical scaffold and an N-degron-recognizing UBR box in the middle. The dimer is assembled as a stable head-to-tail dimer via extensive interactions in the middle helical scaffold region. The tetramer is assembled via SBB2-SBB2 interaction from two face-to-face dimers, forming a large cage with all four catalytic HECT domains facing the central cavity. Importantly, the N-terminal region of one subunit and the HECT of the other form an "intermolecular jaw" in the dimer. Using enzymatic and cellular assays, we showed that the jaw-lining residues are important for function, suggesting that the intermolar jaw functions to recruit ubiquitin loaded E2 to UBR5 for the transthiolation reaction. Further work is needed to understand how oligomerization regulates the UBR5 ligase activity. This work provides a framework for structure-based anticancer drug development against the distinctive HECT E3 ligase and contributes to a growing appreciation of E3 ligase diversity.

biochemistry↗

bioRxiv2022

Naturally occurring Alzheimer's disease in rhesus monkeys

Alzheimers disease (AD) is the most common neurodegenerative disease. To date, its cause is unclear and there are no effective treatments or preventive measures. Despite there are accumulating evidences for the existence of AD pathological hallmarks in the brain of aging rhesus monkeys, it remains a mainstream notion that monkeys do not develop AD naturally. This is an important issue because it will determine how we use monkeys in AD studies. To settle down this issue, a group (n=10) of aged rhesus monkeys 26 years old or above went through a systematic AD screening procedure in this study. Three of these monkeys showed severe memory impairments (SMI) after evaluated with a classic working memory test. Further behavioral testing revealed that the SMI monkeys also exhibited apathy-like behavior, which is another core AD clinical symptom. In addition to the cognitive deficits, two of the three SMI monkeys developed all of the three AD pathological hallmarks, including neurofibrillary tangles, senile plaques and neuronal loss. According to the diagnostic criteria of human AD, the two SMI monkeys were clearly naturally occurring AD monkeys. These results suggest that AD is not a uniquely human disease and monkeys have great potential for the development of much needed etiological AD models, which are vital for better understanding of developmental process of AD and the base of identification of early diagnostic biomarkers and effective therapeutic targets of AD.

neuroscience↗

bioRxiv2022

DNA-encoded library (DEL)-enabled discovery of proximity-inducing small molecules

Molecular glues and bifunctional compounds that induce protein-protein associations provide a powerful and general mechanism to modulate cell circuitry. We sought to develop a platform for the direct discovery of compounds able to induce association of any two pre-selected proteins, using the first bromodomain of BRD4 and the VHL-elongin C-elongin B (VCB) complex as a test system. Leveraging the screening power of DNA-encoded libraries (DELs), we synthesized [~]one million DNA-encoded compounds that possess a VHL-targeting fragment, a variety of connectors, and a diversity element generated by split- and-pool combinatorial chemistry. By screening our DEL against BRD4BD1 in the presence and absence of VCB, we could identify VHL-bound molecules that simultaneously bind BRD4. For highly barcode-enriched library members, ternary complex formation leading to BRD4 degradation was confirmed in cells. Furthermore, a ternary complex crystal structure was obtained for the most enriched library member. Our work provides a foundation for adapting DEL screening to the discovery of proximity-inducing small molecules. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=73 SRC="FIGDIR/small/512184v1_ufig1.gif" ALT="Figure 1"> View larger version (22K): org.highwire.dtl.DTLVardef@43f3c9org.highwire.dtl.DTLVardef@13a1f9eorg.highwire.dtl.DTLVardef@f1d751org.highwire.dtl.DTLVardef@16f3789_HPS_FORMAT_FIGEXP M_FIG C_FIG

pharmacology and toxicology↗

bioRxiv2022

The molecular mechanism of chloramphenicol and thiamphenicol resistance mediated by a novel oxidase CmO in Sphingomonadaceae

Antibiotic resistance mediated by bacterial enzyme inactivation plays a mysterious and crucial role for antibiotic degradation and selection pressure reduction in the environment. The enzymatic inactivation of the antibiotic chloramphenicol (CAP) involves nitro reduction, amide bond hydrolysis and acetylation modification. However, the molecular mechanism of enzymatic oxidation of CAP remains unknown. Here, a novel oxidase gene cmO was identified and confirmed biochemically to catalyze the resistance process through the oxidative inactivation at the side chain C-3 position of CAP and thiamphenicol (TAP) in Sphingomonadaceae. The oxidase CmO is highly conservative in Sphingomonadaceae and shares the highest amino acid homology of 41.05% with the biochemically identified glucose methanol choline (GMC) oxidoreductases. Molecular docking and site-directed mutagenesis analyses demonstrated that CAP was anchored inside the protein pocket of CmO with the hydrogen bonding of key residues glycine (G)99, asparagine (N)518, methionine (M)474 and tyrosine (Y)380. CAP sensitivity test demonstrated that the acetyltransferase and CmO showed higher resistance to CAP as compared with the amide bond-hydrolyzing esterase and nitroreductase. This study provides a better theoretical basis and a novel diagnostic gene for understanding and assessing the fate and resistance risk of CAP and TAP in the environment. ImportanceRising levels of antibiotic resistance undermines ecological and human health as a result of indiscriminate usage of antibiotics. Various resistance mechanisms have been revealed, for instance genes encoding proteins that degrade antibiotics, yet requiring further exploration. In this study, we reported a novel gene encoding an oxidase involved in the inactivation of typical amphenicol antibiotics (chloramphenicol and thiamphenicol), and the molecular mechanism was elucidated. The observation provides novel data to understand capabilities of bacteria to tackle antibiotic stress and suggests complex function of enzymes in the context of antibiotic resistance development and antibiotics removal. The reported gene can be further employed as an indicator to monitor amphenicols fate in the environment, benefiting the risk assessment in this era of antibiotic resistance.

microbiology↗

bioRxiv2022

MATR3-antisense LINE1 RNA meshwork scaffolds higher-order chromatin organization

Long interspersed nuclear elements (LINEs) play essential role in shaping chromatin state, while the factors that cooperate with LINEs and their roles in higher-order chromatin organization remain poorly understood. Here we show that MATR3, a nuclear matrix protein, interplays with antisense LINE1 (AS L1) RNAs to form into a gel-like meshwork via phase-separation, providing a partially dynamic platform for chromatin spatial organization. Either depletion of MATR3 or AS L1 RNAs changes nuclear distribution of each other and leads to chromatin reorganization in the nucleus. After MATR3 depletion, topologically associating domains (TADs) that highly transcribed MATR3-associated AS L1 RNAs showed a decrease on local chromatin interactions. Furthermore, amyotrophic lateral sclerosis (ALS)-associated MATR3 mutants alter biophysical features of the MATR3-AS L1 RNA meshwork and cause chromatin reorganization. Collectively, we revealed an essential role of meshwork formed by nuclear matrix and retrotransposon-derived RNAs in gathering chromatin in the nucleus.

cell biology↗

bioRxiv2022

Acetylation of a fungal effector that translocates host PR1 facilitates virulence

Pathogens utilize a panoply of effectors to manipulate plant defense. However, despite their importance, relatively little is actually known about regulation of these virulence factors. Here, we show that the effector FolSvp1, secreted from fungal pathogen Fusarium oxysporum f. sp. lycopersici (Fol) directly binds and translocates the tomato pathogenesis-related protein1, SlPR1, from the apoplast outside the plasma membrane to the host nucleus via its nuclear localization signal. Relocation of SlPR1 abolishes generation of the defense signaling peptide, CAPE1, from its C-terminus, and as a consequence, facilitates pathogen invasion of plants. The action of FolSvp1 requires covalent modification by acetylation for full virulence in host tomato tissues. The modification is catalyzed by the Fol FolArd1 lysine acetyltransferase prior to secretion. Addition of an acetyl group to one residue, K167, prevents ubiquitination-dependent degradation of FolSvp1 in both Fol and plant cells with different mechanisms, allowing it to function normally in fungal invasion. Either inactivation of FolSvp1 or removal of the acetyl group on K167 leads to impaired pathogenicity of Fol. These findings indicate that acetylation can regulate the stability of effectors of fungal plant pathogens with impact on virulence.

microbiology↗

bioRxiv2022

CIDP: A multi-functional platform for designing CRISPR sgRNAs

Most of sgRNA-design tools can be run under the precondition of the choice of closely-related species. However, it is hard to select an ideal closely-related species, as more and more different species were studied, and this situation was particular seriously in plant studies. Here, we introduced a new software, CRISPR Integrated Design Platform (CIDP), to solve the problem by allowing users to input genomic sequences for designing sgRNAs. The main function of CIDP was to design sgRNAs after building the sgRNA database using the input genomic sequences. Furthermore, in order to minimize the off-target effects, CIDP will search sgRNAs that appear only once across the whole genome on the target sequences. Meanwhile, CIDP set relevant functions to identify shared sgRNAs of a group of genes. Moreover, we also set primer design and sequence extraction functions in CIDP to help users design sgRNAs efficiently.

bioinformatics↗

bioRxiv2022

When two is a crowd - Structural mapping of PEAK pseudokinase interactions identifies 14 3 3 as a molecular switch for PEAK3/Crk signaling.

PEAK pseudokinases regulate cell migration, invasion and proliferation by recruiting key signaling proteins to the cytoskeleton. Despite lacking catalytic activity, alteration in their expression level is associated with several aggressive cancers. Here, we elucidate new molecular details of key PEAK signaling interactions with the adapter proteins CrkII and Grb2 and the scaffold protein 14-3-3. Our findings rationalize why the dimerization of PEAK proteins has a crucial function in signal transduction and provide biophysical and structural data to unravel binding specificity within the PEAK interactome. We identify a conserved high affinity 14-3-3 motif on PEAK3 and demonstrate its role as a molecular switch to regulate CrkII binding. Together, our studies provide a detailed structural snapshot of PEAK interaction networks and further elucidate how PEAK proteins, especially PEAK3, act as dynamic scaffolds that exploit adapter proteins to control signal transduction in cell growth/motility and cancer.

biochemistry↗

bioRxiv2022

Using adversarial networks to extend brain computer interface decoding accuracy over time

Existing intracortical brain computer interfaces (iBCIs) transform neural activity into control signals capable of restoring movement to persons with paralysis. However, the accuracy of the "decoder" at the heart of the iBCI typically degrades over time due to turnover of recorded neurons. To compensate, decoders can be recalibrated, but this requires the user to spend extra time and effort to provide the necessary data, then learn the new dynamics. As the recorded neurons change, one can think of the underlying movement intent signal being expressed in changing coordinates. If a mapping can be computed between the different coordinate systems, it may be possible to stabilize the original decoders mapping from brain to behavior without recalibration. We previously proposed a method based on Generalized Adversarial Networks (GANs), called "Adversarial Domain Adaptation Network" (ADAN), which aligns the distributions of latent signals within underlying low-dimensional neural manifolds. However, ADAN was tested on only a very limited dataset. Here we propose a method based on Cycle-Consistent Adversarial Networks (Cycle-GAN), which aligns the distributions of the full-dimensional neural recordings. We tested both Cycle-GAN and ADAN on data from multiple monkeys and behaviors and compared them to a linear method based on Procrustes Alignment of axes provided by Factor Analysis (PAF). Both GAN-based methods outperformed PAF. Cycle-GAN and ADAN (like PAF) are unsupervised and require little data, making them practical in real life. Overall, Cycle-GAN had the best performance and was easier to train and more robust than ADAN, making it ideal for stabilizing iBCI systems over time. Significance StatementThe inherent instabilities in the neural signals acquired by intracortical microelectrode arrays cause the performance of an intracortical brain computer interface (iBCI) decoder to drop over time, as the movement intent signal must essentially be recorded from neurons representing an ever-changing coordinate system. Here, we address this problem using Generative Adversarial Networks (GANs) to align these coordinates and compare their success to another, recently proposed linear method that uses Factor Analysis and Procrustes alignment. Our proposed methods are fully unsupervised, can be trained quickly, and require remarkably little new data. These methods should give iBCI users access to decoders with unchanging dynamics, and without the need for periodic supervised recalibration.

neuroscience↗