Biology subjects

Samee, M. A. H.

Publications and source records attributed to Samee, M. A. H..

23 records · Page 2Linked to original sources

bioRxiv2021

Mutation rate variations in the human genome are encoded in DNA shape

Single nucleotide mutation rates have critical implications for human evolution and genetic diseases. Accurate modeling of these mutation rates has long remained an open problem since the rates vary substantially across the human genome. A recent model, however, explained much of the variation by considering higher order nucleotide interactions in the local (7-mer) sequence context around mutated nucleotides. Despite this models predictive value, we still lack a biophysically-grounded understanding of genome-wide mutation rate variations. DNA shape features are geometric measurements of DNA structural properties, such as helical twist and tilt, and are known to capture information on interactions between neighboring nucleotides within a local context. Motivated by this characteristic of DNA shape features, we used them to model mutation rates in the human genome. The DNA shape feature based models show up to 15% higher accuracy than the current nucleotide sequence-based models and pinpoint DNA structural properties predictive of mutation rates in the human genome. Further analyzing the mutation rates of individual positions of transcription factor (TF) binding sites in the human genome, we found a strong association between DNA shape and the position-specific mutation rates. The trend holds for hundreds of TFs and is even stronger in evolutionarily conserved regions. To our knowledge, this is the first attempt that demonstrates the structural underpinnings of nucleotide mutations in the human genome and lays the groundwork for future studies to incorporate DNA shape information in modeling genetic variations.

genomics↗

bioRxiv2020

Integrative Analysis of Spatial and Single-Cell Transcriptomics Reveals Principles of Tissue Organization and Intercellular Communication in Mouse Olfactory Bulb

Intercellular communication and spatial organization of cells are two critical aspects of a tissues function. Understanding these aspects requires integrating data from single-cell RNA-Seq (scRNA-seq) and spatial transcriptomics (ST), the two cutting edge technologies that offer complementary insights into tissue composition, architecture, and function. Integrating these data types is non-trivial since they differ widely in the number of profiled genes and often do not share marker genes for given cell-types. We developed STANN, a neural network model that overcomes these methodological challenges. Given ST and scRNA-seq data of a tissue, STANN models cell-types in the scRNA-seq dataset from the genes that are profiled by both ST and scRNA-seq. The trained STANN model then assigns cell-types to the ST dataset. We apply STANN to assign cell-types in a recent ST dataset (SeqFISH+) of mouse olfactory bulb (MOB). Our analysis of STANNs assigned cell-types revealed principles of tissue architecture and intercellular communication at unprecedented detail. We find that cell-type compositions are disproportionate in the tissue, yet their relative proportions are spatially consistent within individual morphological layers. Surprisingly, within a morphological layer, there is a high spatial variation in cell-type colocalization patterns and intercellular communication mechanisms. Our analysis suggests that spatially localized gene regulatory networks may account for such variability in intercellular communication mechanisms.

systems biology↗

bioRxiv2020

Multinomial Convolutions for Joint Modeling of Sequence Motifs and Enhancer Activities

Massively parallel reporter assays (MPRAs) have enabled the study of transcriptional regulatory mechanisms at an unprecedented scale and with high quantitative resolution. However, this realm lacks models that can discover sequence-specific signals de novo from the data and integrate them in a mechanistic way. We present MuSeAM (Multinomial CNNs for Sequence Activity Modeling), a convolutional neural network that overcomes this gap. MuSeAM utilizes multinomial convolutions that directly model sequence-specific motifs of protein-DNA binding. We demonstrate that MuSeAM fits MPRA data with high accuracy and generalizes over other tasks such as predicting chromatin accessibility and prioritizing potentially functional variants.

systems biology↗

bioRxiv2018

Massively parallel dissection of human accelerated regions in human and chimpanzee neural progenitors

Using machine learning (ML), we interrogated the function of all human-chimpanzee variants in 2,645 Human Accelerated Regions (HARs), some of the fastest evolving regions of the human genome. We predicted that 43% of HARs have variants with large opposing effects on chromatin state and 14% on neurodevelopmental enhancer activity. This pattern, consistent with compensatory evolution, was confirmed using massively parallel reporter assays in human and chimpanzee neural progenitor cells. The species-specific enhancer activity of assayed HARs was accurately predicted from the presence and absence of transcription factor footprints in each species. Despite these striking cis effects, activity of a given HAR sequence was nearly identical in human and chimpanzee cells. These findings suggest that HARs did not evolve to compensate for changes in the trans environment but instead altered their ability to bind factors present in both species. Thus, ML prioritized variants with functional effects on human neurodevelopment and revealed an unexpected reason why HARs may have evolved so rapidly.

evolutionary biology↗

bioRxiv2017

Transcription Factors Recognize DNA Shape Without Nucleotide Recognition

We hypothesized that transcription factors (TFs) recognize DNA shape without nucleotide sequence recognition. Motivating an independent role for shape, many TF binding sites lack a sequence-motif, DNA shape adds specificity to sequence-motifs, and different sequences can encode similar shapes. We therefore asked if binding sites of a TF are enriched for specific patterns of DNA shape-features, e.g., helical twist. We developed ShapeMF, which discovers these shape-motifs de novo without taking sequence information into account. We find that most TFs assayed in ENCODE have shape-motifs and bind regulatory regions recognizing shape-motifs in the absence of sequence-motifs. When shape- and sequence-recognition co-occur, the two types of motifs can be overlapping, flanking, or separated by consistent spacing. Shape-motifs are prevalent in regions co-bound by multiple TFs. Finally, TFs with identical sequence motifs have different shape-motifs, explaining their binding at distinct locations. These results establish shape-motifs as drivers of TF-DNA recognition complementary to sequence-motifs.

systems biology↗