Biology subjects

Heinemann, U.

Publications and source records attributed to Heinemann, U..

2 recordsLinked to original sources

bioRxiv2026

Quantification of the effects of single nucleotide variants in NKX2.1 transcription factor binding sites

Transcription factors recognise and bind specific DNA sequence patterns in promoters and enhancers thereby regulating gene expression. Variations in the DNA sequence of transcription factor binding sites (TFBSs) can alter gene regulation and may disrupt development. The transcription factor NKX2.1 is a crucial regulator of thyroid, lung, and neural development. Mutations in its coding gene NKX2-1 may cause choreoathetosis and congenital hypothyroidism with or without pulmonary dysfunction (CAHTP, OMIM #610978). Most genetically solved patients carry mutations in the coding regions of NKX2-1 that affect DNA binding, while the majority of patients with CAHTP-like symptoms do not carry mutations in the NKX2-1 coding sequence. We hypothesise that variations in the DNA-sequence at promoter or enhancer sites to which the transcription factor NKX2.1 binds could cause disease as well. We employed EMSA-seq to quantify the effects of genetic variation on NKX2.1 binding strength and used this data to train neural network models to forecast the influence of DNA variation on NKX2.1 binding. We validated our models using microscale thermophoresis, X-ray crystallography, and publicly available ChIP-seq data sets. The neural networks were able to detect TFBSs in ChIP-seq data and can thus be used to evaluate whole genome sequencing data of CAHTP-patients in order to prioritise potential disease-causing variants in regulatory elements. Graphical abstract O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=80 SRC="FIGDIR/small/708450v2_ufig1.gif" ALT="Figure 1"> View larger version (31K): org.highwire.dtl.DTLVardef@167cedeorg.highwire.dtl.DTLVardef@3e5291org.highwire.dtl.DTLVardef@19eb7f9org.highwire.dtl.DTLVardef@1404057_HPS_FORMAT_FIGEXP M_FIG C_FIG

neuroscience↗

bioRxiv2022

Discovery of a non-canonical GRHL1 binding motif using deep convolutional and recurrent neural networks

Transcription factors regulate gene expression by binding to transcription factor binding sites (TFBSs). Most models for predicting TFBSs are based on position weight matrices (PWMs), which require a specific motif to be present in the DNA sequence and do not consider interdependencies of nucleotides. Novel approaches such as Transcription Factor Flexible Models or recurrent neural networks consequently provide higher accuracies. However, it is unclear whether such approaches can uncover novel non-canonical, hitherto unexpected TFBSs relevant to human transcriptional regulation. In this study, we trained a convolutional recurrent neural network with HT-SELEX data for GRHL1 binding and applied it to a set of GRHL1 binding sites obtained from ChIP-Seq experiments from human cells. We identified 46 non-canonical GRHL1 binding sites, which were not found by a conventional PWM approach. Unexpectedly, some of the newly predicted binding sequences lacked the CNNG core motif, so far considered obligatory for GRHL1 binding. Using isothermal titration calorimetry, we experimentally confirmed binding between the GRHL1-DNA binding domain and predicted GRHL1 binding sites, including a non-canonical GRHL1 binding site. Mutagenesis of individual nucleotides revealed a correlation between predicted binding strength and experimentally validated binding affinity across representative sequences. This correlation was neither observed with a PWM-based nor another deep learning approach. Our results show that convolutional recurrent neural networks may uncover unanticipated binding sites and facilitate quantitative transcription factor binding predictions.

bioinformatics↗