BI project

Deep learning in protein anomalies

Convolutional neural networks (CNNs) are well known for their ability to leverage spatial and temporal structure. They can accurately infer protein functions or predict small molecules’ bioactivity in ligand-protein interactions. Nevertheless, several technical issues limit their wider adoption in structural biology. Here we address some of the most challenging of them.

Graphical application for clinical trial design in oncohematology

Hematopoietic stem cell transplantation (HSCT) is a dangerous procedure with many possible complications, so it is reserved for patients with cancers of the blood or bone marrow. The recipient's immune system is usually destroyed with radiation or chemotherapy before the transplantation. The major complications of HSCT are infections and graft-versus-host disease. Such transplantation is relatively common in R.M. Gorbacheva Memorial Research Institute of Oncology, Hematology and Transplantation in 2015.

The optimization of de novo transcriptome assembly strategies

De novo transcriptome assembly could provide essential biological information about organism of interest, especially when full genome assembly and annotation are not feasible due to large genome size, ploidy, or for other reasons.

Statistical analysis of neural spike trains for estimation of functional differences in subcortical structures of human brain

In this project we studied spontaneous single unit activity in subthalamic nucleus (STN) of Parkinson’s disease patients. Microelectrode recording (MER) was performed during deep brain stimulation (DBS) stereotactic neurosurgery. The aim of the current study is to find statistical properties of neural spike trains in STN that differ anesthetized and awake states. In our research we analyzed spontaneous activity of 114 cells and 183 cells of 8 Parkinson’s disease patients in anesthetized and awake state respectively.

Response of the cells with suppressed chromatin repression system to osmotic stress

DNA methylation is implicated in transcription regulation, mostly by repressing initiation of transcription. Such repression often appears as a result of a repressive complex binding to methylated genomic regions. Kaiso, a transcription factor that binds methylated DNA sequence, can form a complex with N-CoR causing histone deacetylation and, therefore, repression of transcription. Osmotic stress has been shown to result in Kaiso translocation to inner surface of nuclear membrane reducing its DNA binding.

Clinical Somatic Mutant Caller: detection of clinically relevant somatic mutations in NGS data.

Since the emergence of high-throughput genomic sequencing technologies, a huge volume of data has been accumulated on somatic mutations in cancer. Only limited number of mutations have been associated with a known clinical prognosis and can be used to determine accurate diagnosis and targeted treatment. These mutations are located in cancer gene “hot spots” and usually interfere with oncogene functionality.

Intestinal microbiota in health and disease

Intestinal microbiota is constantly changing depending on various conditions. It is a well known fact that in different diseases shifts in intestinal microbiota are observed. But treatment of such dysbiotic states remains a big challenge.

The aim of this research is to find the most efficient autoprobiotic for the treatment of patients with Irritable Bowel Syndrome (IBS). It can be achieved by the analysis of changes in microbiota towards its normalization during treatment.

Text-mining based retrieving of omics pipelines

In this work we developed a method for automatic extraction of proteomics pipelines from scientific papers. The objective was to figure out which proteomics tools are incorporated together, with respect to data formats, frequency of use, and efficacy of the data analysis performed.

Improving NLP in molecular biology

Natural language processing (NLP) techniques have seen a steady rise in complexity and popularity, becoming increasingly important in text-mining applications in data-intensive areas, such as computational biology. Unfortunately, many novel methods, based on neural networks and unsupervised deep learning, require adaptation to be efficiently applied to molecular-biological texts.

Human genome variations analysis — 2

Cardiomyopathies are a heterogeneous group of diseases that are associated with functional disturbances of heart muscle. Even though genetic predisposition is frequently observed in patients' anamnesis, a whole set of genes responsible for cardiomyopathy emergence is not determined yet.

46 target genes important in the research of inherited cardiomyopathies were sequenced by TruSight Cardiomyopathy Illumina panel. The data was obtained from 3 families of 3-4 members with at least one member affected by the disease. 


Subscribe to RSS - BI project