TY - JOUR A1 - Dirk, Robin A1 - Fischer, Jonas L. A1 - Schardt, Simon A1 - Ankenbrand, Markus J. A1 - Fischer, Sabine C. T1 - Recognition and reconstruction of cell differentiation patterns with deep learning JF - PLoS Computational Biology N2 - Abstract Cell lineage decisions occur in three-dimensional spatial patterns that are difficult to identify by eye. There is an ongoing effort to replicate such patterns using mathematical modeling. One approach uses long ranging cell-cell communication to replicate common spatial arrangements like checkerboard and engulfing patterns. In this model, the cell-cell communication has been implemented as a signal that disperses throughout the tissue. On the other hand, machine learning models have been developed for pattern recognition and pattern reconstruction tasks. We combined synthetic data generated by the mathematical model with spatial summary statistics and deep learning algorithms to recognize and reconstruct cell fate patterns in organoids of mouse embryonic stem cells. Application of Moran’s index and pair correlation functions for in vitro and synthetic data from the model showed local clustering and radial segregation. To assess the patterns as a whole, a graph neural network was developed and trained on synthetic data from the model. Application to in vitro data predicted a low signal dispersion value. To test this result, we implemented a multilayer perceptron for the prediction of a given cell fate based on the fates of the neighboring cells. The results show a 70% accuracy of cell fate imputation based on the nine nearest neighbors of a cell. Overall, our approach combines deep learning with mathematical modeling to link cell fate patterns with potential underlying mechanisms. Author summary Mammalian embryo development relies on organized differentiation of stem cells into different lineages. Particularly at the early stages of embryogenesis, cells of different fates form three-dimensional spatial patterns that are difficult to identify by eye. Pattern quantification and mathematical modeling have produced first insights into potential mechanisms for the cell fate arrangements. However, these approaches have relied on classifications of the patterns such as inside-out or random, or used summary statistics such as pair correlation functions or cluster radii. Deep neural networks allow characterizing patterns directly. Since the tissue context can be readily reproduced by a graph, we implemented a graph neural network to characterize the patterns of embryonic stem cell organoids as a whole. In addition, we implemented a multilayer perceptron model to reconstruct the fate of a given cell based on its neighbors. To train and test the models, we used synthetic data generated by our mathematical model for cell-cell communication. This interplay of deep learning and mathematical modeling in combination with summary statistics allowed us to identify a potential mechanism for cell fate determination in mouse embryonic stem cells. Our results agree with a mechanism with a dispersion of the intercellular signal that links a cell’s fate to those of the local neighborhood. KW - recognition KW - reconstruction KW - cell differentiation patterns KW - deep learning KW - mouse embryonic stem cells KW - multilayer perceptron model Y1 - 2023 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-350167 VL - 19 IS - 10 ER - TY - JOUR A1 - Bangalore, Disha M. A1 - Heil, Hannah S. A1 - Mehringer, Christian F. A1 - Hirsch, Lisa A1 - Hemmen, Katharina A1 - Heinze, Katrin G. A1 - Tessmer, Ingrid T1 - Automated AFM analysis of DNA bending reveals initial lesion sensing strategies of DNA glycosylases JF - Scientific Reports N2 - Base excision repair is the dominant DNA repair pathway of chemical modifications such as deamination, oxidation, or alkylation of DNA bases, which endanger genome integrity due to their high mutagenic potential. Detection and excision of these base lesions is achieved by DNA glycosylases. To investigate the remarkably high efficiency in target site search and recognition by these enzymes, we applied single molecule atomic force microscopy (AFM) imaging to a range of glycosylases with structurally different target lesions. Using a novel, automated, unbiased, high-throughput analysis approach, we were able to resolve subtly different conformational states of these glycosylases during DNA lesion search. Our results lend support to a model of enhanced lesion search efficiency through initial lesion detection based on altered mechanical properties at lesions. Furthermore, its enhanced sensitivity and easy applicability also to other systems recommend our novel analysis tool for investigations of diverse, fundamental biological interactions. KW - atomic-force microscopy KW - base pairs KW - molecular structure KW - crystal structure KW - structural basis KW - repair KW - recognition KW - 8-oxoguanine KW - thymine KW - mismatches Y1 - 2020 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-231338 VL - 10 ER - TY - JOUR A1 - Metje-Sprink, Janina A1 - Groffmann, Johannes A1 - Neumann, Piotr A1 - Barg-Kues, Brigitte A1 - Ficner, Ralf A1 - Kühnel, Karin A1 - Schalk, Amanda M. A1 - Binotti, Beyenech T1 - Crystal structure of the Rab33B/Atg16L1 effector complex JF - Scientific Reports N2 - The Atg12-Atg5/Atg16L1 complex is recruited by WIPI2b to the site of autophagosome formation. Atg16L1 is an effector of the Golgi resident GTPase Rab33B. Here we identified a minimal stable complex of murine Rab33B(30-202) Q92L and Atg16L1(153-210). Atg16L1(153-210) comprises the C-terminal part of the Atg16L1 coiled-coil domain. We have determined the crystal structure of the Rab33B Q92L/Atg16L1(153-210) effector complex at 3.47 angstrom resolution. This structure reveals that two Rab33B molecules bind to the diverging alpha -helices of the dimeric Atg16L1 coiled-coil domain. We mutated Atg16L1 and Rab33B interface residues and found that they disrupt complex formation in pull-down assays and cellular co-localization studies. The Rab33B binding site of Atg16L1 comprises 20 residues and immediately precedes the WIPI2b binding site. Rab33B mutations that abolish Atg16L binding also abrogate Rab33B association with the Golgi stacks. Atg16L1 mutants that are defective in Rab33B binding still co-localize with WIPI2b in vivo. The close proximity of the Rab33B and WIPI2b binding sites might facilitate the recruitment of Rab33B containing vesicles to provide a source of lipids during autophagosome biogenesis. KW - autophagosome formation KW - ATG12-ATG5 conjugate KW - LC3 lipidation KW - binding sites KW - ATG proteins KW - RAB GTPases KW - family KW - membrane KW - recognition KW - proppins Y1 - 2020 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-230396 VL - 10 ER - TY - JOUR A1 - Peissert, Stefan A1 - Sauer, Florian A1 - Grabarczyk, Daniel B. A1 - Braun, Cathy A1 - Sander, Gudrun A1 - Poterszman, Arnaud A1 - Egly, Jean-Marc A1 - Kuper, Jochen A1 - Kisker, Caroline T1 - In TFIIH the Arch domain of XPD is mechanistically essential for transcription and DNA repair JF - Nature Communications N2 - The XPD helicase is a central component of the general transcription factor TFIIH which plays major roles in transcription and nucleotide excision repair (NER). Here we present the high-resolution crystal structure of the Arch domain of XPD with its interaction partner MAT1, a central component of the CDK activating kinase complex. The analysis of the interface led to the identification of amino acid residues that are crucial for the MAT1-XPD interaction. More importantly, mutagenesis of the Arch domain revealed that these residues are essential for the regulation of (i) NER activity by either impairing XPD helicase activity or the interaction of XPD with XPG; (ii) the phosphorylation of the RNA polymerase II and RNA synthesis. Our results reveal how MAT1 shields these functionally important residues thereby providing insights into how XPD is regulated by MAT1 and defining the Arch domain as a major mechanistic player within the XPD scaffold. KW - nucleotide excision repair KW - nuclear receptors KW - helicase KW - transactivation KW - fluorescence KW - recognition KW - subunit KW - binding KW - sulfur KW - kinease Y1 - 2020 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-229857 VL - 11 IS - 1 ER - TY - JOUR A1 - Gruss, L. Forest A1 - Wieser, Matthias J. A1 - Schweinberger, Stefan R. A1 - Keil, Andreas T1 - Face-evoked steady-state visual potentials: effects of presentation rate and face inversion JF - Frontiers in Human Neuroscience N2 - Face processing can be explored using electrophysiological methods. Research with event-related potentials has demonstrated the so-called face inversion effect, in which the N170 component is enhanced in amplitude and latency to inverted, compared to upright, faces. The present study explored the extent to which repetitive lower-level visual cortical engagement, reflected in flicker steady-state visual evoked potentials (ssVEPs), shows similar amplitude enhancement to face inversion. We also asked if inversion-related ssVEP modulation would be dependent on the stimulation rate at which upright and inverted faces were flickered. To this end, multiple tagging frequencies were used (5, 10, 15, and 20 Hz) across two studies (n=21, n=18). Results showed that amplitude enhancement of the ssVEP for inverted faces was found solely at higher stimulation frequencies (15 and 20 Hz). By contrast, lower frequency ssVEPs did not show this inversion effect. These findings suggest that stimulation frequency affects the sensitivity of ssVEPs to face inversion. KW - N170 KW - upside-down faces KW - selective attention KW - spatial attention KW - cortex KW - perception KW - recognition KW - brain KW - FMRI KW - area Y1 - 2012 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-134399 VL - 6 IS - 316 ER - TY - JOUR A1 - Jin, Jing A1 - Allison, Brendan Z. A1 - Kaufmann, Tobias A1 - Kübler, Andrea A1 - Zhang, Yu A1 - Wang, Xingyu A1 - Cichocki, Andrzej T1 - The Changing Face of P300 BCIs: A Comparison of Stimulus Changes in a P300 BCI Involving Faces, Emotion, and Movement JF - PLoS One N2 - Background: One of the most common types of brain-computer interfaces (BCIs) is called a P300 BCI, since it relies on the P300 and other event-related potentials (ERPs). In the canonical P300 BCI approach, items on a monitor flash briefly to elicit the necessary ERPs. Very recent work has shown that this approach may yield lower performance than alternate paradigms in which the items do not flash but instead change in other ways, such as moving, changing colour or changing to characters overlaid with faces. Methodology/Principal Findings: The present study sought to extend this research direction by parametrically comparing different ways to change items in a P300 BCI. Healthy subjects used a P300 BCI across six different conditions. Three conditions were similar to our prior work, providing the first direct comparison of characters flashing, moving, and changing to faces. Three new conditions also explored facial motion and emotional expression. The six conditions were compared across objective measures such as classification accuracy and bit rate as well as subjective measures such as perceived difficulty. In line with recent studies, our results indicated that the character flash condition resulted in the lowest accuracy and bit rate. All four face conditions (mean accuracy >91%) yielded significantly better performance than the flash condition (mean accuracy = 75%). Conclusions/Significance: Objective results reaffirmed that the face paradigm is superior to the canonical flash approach that has dominated P300 BCIs for over 20 years. The subjective reports indicated that the conditions that yielded better performance were not considered especially burdensome. Therefore, although further work is needed to identify which face paradigm is best, it is clear that the canonical flash approach should be replaced with a face paradigm when aiming at increasing bit rate. However, the face paradigm has to be further explored with practical applications particularly with locked-in patients. KW - ERPS KW - communication KW - TO-target interval KW - visual-evoked potentials KW - brain-computer-interface KW - recognition KW - amplitude KW - paradigm KW - systems Y1 - 2012 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-134173 VL - 7 IS - 11 ER - TY - JOUR A1 - Kincaid, Rodney P. A1 - Chen, Yating A1 - Cox, Jennifer E. A1 - Rethwilm, Axel A1 - Sullivan, Christopher S. T1 - Noncanonical MicroRNA (miRNA) Biogenesis Gives Rise to Retroviral Mimics of Lymphoproliferative and Immunosuppressive Host miRNAs JF - mBio N2 - MicroRNAs (miRNAs) play regulatory roles in diverse processes in both eukaryotic hosts and their viruses, yet fundamental questions remain about which viruses code for miRNAs and the functions that they serve. Simian foamy viruses (SFVs) of Old World monkeys and apes can zoonotically infect humans and, by ill-defined mechanisms, take up lifelong infections in their hosts. Here, we report that SFVs encode multiple miRNAs via a noncanonical mode of biogenesis. The primary SFV miRNA transcripts (pri-miRNAs) are transcribed by RNA polymerase III (RNAP III) and take multiple forms, including some that are cleaved by Drosha. However, these miRNAs are generated in a context-dependent fashion, as longer RNAP II transcripts spanning this region are resistant to Drosha cleavage. This suggests that the virus may avoid any fitness penalty that could be associated with viral genome/transcript cleavage. Two SFV miRNAs share sequence similarity and functionality with notable host miRNAs, the lymphoproliferative miRNA miR-155 and the innate immunity suppressor miR-132. These results have important implications regarding foamy virus biology, viral miRNAs, and the development of retroviral-based vectors. IMPORTANCE Fundamental questions remain about which viruses encode miRNAs and their associated functions. Currently, few natural viruses with RNA genomes have been reported to encode miRNAs. Simian foamy viruses are retroviruses that are prevalent in nonhuman host populations, and some can zoonotically infect humans who hunt primates or work as animal caretakers. We identify a cluster of miRNAs encoded by SFV. Characterization of these miRNAs reveals evolutionarily conserved, unconventional mechanisms to generate small RNAs. Several SFV miRNAs share sequence similarity and functionality with host miRNAs, including the oncogenic miRNA miR-155 and innate immunity suppressor miR-132. Strikingly, unrelated herpesviruses also tap into one or both of these same regulatory pathways, implying relevance to a broad range of viruses. These findings provide new insights with respect to foamy virus biology and vectorology. KW - MIR-155 KW - simian foamy viruses KW - long terminal repeat KW - viral microRNAs KW - RNA KW - infection KW - herpesvirus KW - recognition KW - prediction KW - ortholog Y1 - 2014 U6 - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:20-opus-117216 SN - 2150-7511 VL - 5 IS - 2 ER -