imagine (Institute of molecular genetics and genetic engineering)
Not a member yet
3088 research outputs found
Sort by
Towards a linearly organised embedding space of biological networks
No full textLow-dimensional embeddings are a cornerstone in the modelling and analysis of complex
biological networks. Embedding biological networks is challenging, as it involves capturing
both structural (topological) and semantic information of a graph (i.e., node labels).
Typically, nodes with the same label are in the same dense subgraph (neighborhood-based
similarity), but it has been shown that similarly annotated nodes can be in different network
neighborhoods while having similar wiring patterns (topological similarity). However, current
network embedding algorithms do not preserve both types of similarity, which limits the
information preserved in the embedding space. Moreover, most existing approaches for
mining network embedding spaces rely on computationally intensive machine learning
systems to facilitate downstream analysis tasks. On the other hand, word embedding spaces
capture semantic relationships linearly, allowing for information retrieval using simple linear
operations on word embedding vectors.
In our work, following the NLP paradigm, we introduce novel random-walk-based embeddings
that allow mining biological knowledge directly from the embedding space. Namely, we
introduce embeddings that locate close in the space genes that have similar biological
functions (either topological or neighborhood-based similar nodes). We exploit this property
to predict genes participating in protein complexes and to identify cancer-related genes
based on the cosine similarities between the vector representations of the genes. We also go
beyond embeddings that preserve one type of similarity by introducing novel graphlet-based
representations of the networks that simultaneously capture topological and neighborhood
membership information. We use all the different network representations to assess
whether it is an intrinsic property in the structure of the data (input matrix representation)
that yield embedding spaces that enable downstream analysis tasks via simple linear
operations. Using nine multi-label biological networks and seven single label networks that
are commonly used in machine learning studies, we demonstrate that the more homophilic
the network matrix representation, the more linearly organized the corresponding network
embedding space, and thus, the better the downstream analysis results. Our results suggest
that our new graphlet-based methodologies embed networks into linear spaces, allowing
for better mining of the networks and alleviating the need for computational-intensive ML
models.Book of abstracts: 5th Belgrade Bioinformatics Conference, Serbia, Belgrade,17-20 june 2024
DCS: from Reading Genome to Understanding Life
No full textThe Central Dogma, initially proposed by Francis Crick in 1958, delineates the essential
flow of genetic information within living organisms. It succinctly states that DNA directs
RNA, which in turn guides protein synthesis. Consequently, understanding the genome is
pivotal for comprehending life itself.
DNA Sequencing (D) has emerged as a critical tool in unraveling the mysteries of life.
Leveraging cutting-edge platforms like DNB-based high-throughput sequencing, we have
amassed vast genomic data. These technological breakthroughs have illuminated various
aspects of biology. However, even armed with genomic information, we grapple with a
fundamental question: How does the same genomic blueprint yield diverse cell types?
With Single-cell sequencing technology (C), sequencing individual cells, we’ve gained
insights into how the same set of genome orchestrates the intricate dance of cellular
forms and functions.
Recent breakthroughs in spatial transcripts (S), particularly techniques like Stereo-seq, to
study single cells transcriptome as well as other omics across time and space dimensions.
These approaches seek to address critical questions about genome regulation, its impact
on cellular diversity, and how these processes influence life phenomena, including aging
and disease. The talk will introduce our recent progress of these DCS technologies and
application in different biology research projects.Book of abstracts: 5th Belgrade Bioinformatics Conference, Serbia, Belgrade,17-20 june 2024
MONFIT: Multi-omics factorization-based integration of time-series data sheds light on Parkinson’s disease
No full textParkinson’s disease (PD) is a severe and complex multifactorial neurodegenerative disease whose
elusive pathophysiology prevents the development of curative treatments. Studying PD using
longitudinal multi-omics data is a promising approach to identifying its mechanisms of etiology and
progression. However, heterogeneous data require new analysis frameworks that can utilize the
complementary information captured by diverse data types and further the understanding of PD
across biological entities and processes.
We present MONFIT, a holistic analysis pipeline that integrates and mines time-series single-cell
RNA-sequencing data of disease and control cell lines, along with bulk proteomics and metabolomics
data, by non-negative matrix tri-factorization, hence enabling prior knowledge integration from
molecular networks. MONIFT first integrates (fuses) time-point-specific data, producing timepoint-
specific gene embeddings, which it then collectively mines across time points.
We apply MONFIT to longitudinal, multi-omics data of PD and control cells obtained from patientderived
induced pluripotent stem cells that were differentiated into dopaminergic neurons. We
predict 123 genes related to PD, which we validate by network analysis to be specific to the PINK1
mutation causing PD. We investigate the top 30 gene predictions and propose five novel PD gene
candidates: CENPF, CRABP1, TOP2A, TMSB10, and NASP. In addition, we emphasize molecular
pathways that play important roles in PD pathology and suggest new intervention opportunities
by drug repurposing. We demonstrate that MONFIT goes beyond standard differential analysis
approaches of single-omics data by predicting PD-associated genes that would otherwise elude
discovery. MONFIT is a generic method and can be modified to accommodate data from tissue
samples and other multi-omics data types.Book of abstracts: 5th Belgrade Bioinformatics Conference, Serbia, Belgrade,17-20 june 2024
Intrinsic disorder of proteins associated with diseases
No full textNumerous publicly accessible databases include variously formatted information regarding
the relationship between genes and diseases. This work expedites their use by integrating
them into one standardised database – Integrated Gene Disease Database. IGDD currently
has more than 400,000 rows incorporating gene-disease associations from the following
sources: DisGeNet, COSMIC, HumsaVar, Orphanet, ClinVar, HPO, DISEASES. Its features
include: gene symbol and IDs, UniProt ID, disease name, Disease Ontology ID. Disease
Ontology was chosen since it offers a wide range of possibilities in terms of disease
exploration.
IGDD was further enriched with information on the disorder of the proteins encoded by the
genes associated with diseases since many lack a fixed and well-defined three-dimensional
structure. That fact may be linked with the disease-causing mechanisms, so it is an important
feature of a protein. Several disorder measures were used, based both on the sequence
profiling and the advanced statistical methods: amino acid profiles, charge-hydropathy (CH)
prediction, PONDR family (VL-XT, VSL2), IUPred family (long, short, ANCHOR), FuzDrop.
This work focuses on the following question: is there any relationship between certain
diseases or their groups and the level of disorder of proteins related to them? With that in
mind, no correlation was found between any considered disorder measure and the number
of diseases that proteins are related to. There was neither a correlation between the depth
of diseases in the ontology and the disorder of the related proteins. Additionally, no obvious
regularity was noticed when it comes to the disorder of proteins grouped by diseases they
are related to. Both ordered and disordered proteins were equally found in all parts of the
ontology.
Regardless of the results of this research, IGDD can nevertheless be considered a valuable
resource for future data analysis and further investigation of gene-disease associations. Its
detailed features and a large number of relations open the path for many types of studies.Book of abstracts: 5th Belgrade Bioinformatics Conference, Serbia, Belgrade,17-20 june 2024
Unsupervised domain adaptation methods for cross-species transfer of regulatory code signals
No full textDue to advances in NGS technologies whole-genome maps of various functional genomic
elements were generated for a dozen of species, however experiments are still expensive
and are not available for many species of interest. Deep learning methods became the
state-of-the-art computational methods to analyze the available data, but the focus is
often only on the species studied. Here we take advantage of the progresses in Transfer
Learning in the area of Unsupervised Domain Adaption (UDA) and tested nine UDA
methods for prediction of regulatory code signals for genomes of other species. We tested
each deep learning implementation by training the model on experimental data from one
species, then refined the model using the genome sequence of the target species for which
we wanted to make predictions. Among nine tested domain adaptation architectures
non-adversarial methods Minimum Class Confusion (MCC) and Deep Adaptation Network
(DAN) significantly outperformed others. Conditional Domain Adversarial Network (CDAN)
appeared as the third best architecture. Here we provide an empirical assessment of each
approach using real world data. The different approaches were tested on ChIP-seq data
for transcription factor binding sites and histone marks on human and mouse genomes,
but is generalizable to any cross-species transfer of interest. We tested the efficiency of
each method using species where experimental data was available for both. The results
allows us to assess how well each implementation will work for species for which only
limited experimental data is available and will inform the design of future experiments in
these understudied organisms. Overall, our results proved the validity of UDA methods
for generation of missing experimental data for histone marks and transcription factor
binding sites in various genomes and highlights how robust the various approaches are to
data that is incomplete, noisy and susceptible to analytic bias.Book of abstracts: 5th Belgrade Bioinformatics Conference, Serbia, Belgrade,17-20 june 2024
Enzyme-based strategies for inhibiting medically relevant biofilms: applications in healthcare and food safety
No full textBiofilms are complex communities of micro-
organisms enclosed in a self-produced matrix of extra-
cellular polymeric substances consisting of polysac-
charides, proteins, lipids and extracellular DNA. These
biofilms can form on a variety of surfaces and pose a
major challenge in both the medical and industrial sec-
tors as they can harbor pathogenic microorganisms. In
healthcare, biofilms form on hospital surfaces, medical
devices, patient tissue and implants and contribute to
persistent infections that are difficult to treat. In the
food industry, biofilms on processing equipment and
food matrices can also lead to contamination and pose
a serious threat to public health through foodborne
diseases. The bacteria embedded in biofilms are much
more tolerant to antimicrobial treatments than their
planktonic counterparts, necessitating the develop-
ment of new strategies to combat biofilm-associated
infections and contamination. As biofilms mature,
they become even more resistant to conventional
treatments, making prevention strategies particularly
important. This review focuses on enzyme-based strat-
egies that have been developed over the last decade
to inhibit biofilm formation. Key approaches such as
disruption of microbial signaling pathways and deg-
radation of biofilm matrix components are highlight-
ed, offering promising ways to prevent biofilm-related
problems in both medicine and industry
INFUSION-RELATED REACTIONS TO RITUXIMAB BIOSIMILARS IN PATIENTS WITH CHRONIC B-LYMPHOPROLIFERATIVE DISORDERS: SINGLE CENTER EXPERIENCE
No full textUvod: Infuzione reakcije (IR) predstavljaju česte komplikacije primene rituksimaba,
koje se uglavnom uspešno prevazilaze. U literaturi nismo pronašli poređenje
različitih biosimilara rituksimaba u pogledu IR.
Cilj: Sagledati različite aspekte IR na biosimilare rituksimaba, Rixathon® i Blitzima
® kojima su lečeni pacijenti sa hroničnim B-limfoproliferativnim bolestima
(hBLPB) na Klinici za hematologiju UKCS 2020–2022. godine.
Metode: Iz medicinske dokumentacije su izdvojeni podaci 200 konsekutivnih
pacijenata sa hBLPB koji su lečeni imunohemioterapijom u našoj ustanovi. Svi
pacijenti su primili standardnu premedikaciju pre primene rituksimaba.
Rezultati: IR su se manifestovale kod 24% pacijenata. Učestalost IR nije bila
povezana sa polom, starosnom dobi, stadijumom, lokalizacijom (nodalna/ekstranodalna),
B simptomima, „bulky” masom, performans statusom, tipom hemioterapije
(HT) koja je primenjivana uz rituksimab niti sa vremenom primene
rituksimaba u odnosu na ciklus HT (od I ciklusa vs. od II+ ciklusa). Pacijenti sa
difuznim B krupnoćelijskim limfomom (DBKL) su ispoljili značajno manji broj IRR
u poređenju sa drugima (p = 0,033), u prvom redu hroničnom limfocitnom leukemijom
(15% vs. 34%; p = 0,045). Oni koji su lečeni u drugoj ili kasnijim terapijskim
linijama češće su imali IR (p = 0,0233).
Zaključak: Nije uočena značajna razlika u učestalosti IR između različitih biosimilara.
Pacijenti sa DBKL i oni koji su primali rituksimab u prvoj liniji, ređe su
ispoljili IR u odnosu na pacijente sa drugim tipovima hBLPB, odnosno lečene u
drugoj ili kasnijim linijama. Incidenca IR kod pacijenata koji su primali različite
vrste biosimilara rituksimaba nije se značajno razlikovala. Arterijska hipertenzija,
dijabetes melitus, koronarna bolest i pušenje nisu uticali na incidencu IR.Introduction: Infusion-related reactions (IRRs) represent frequent, although
manageable complications of rituximab administration. To the best of our
knowledge, no comparisons between different biosimilars in this setting have
been published to date.
Objective: To assess different aspects of IRRs to rituximab biosimilars Rixathon®
and Blitzima® administered to patients with chronic B-lymphoproliferative disorders
(cBLPDs) treated at Clinic for Hematology, UCCS, from 2020 to 2022.
Methods: We assessed medical records of 200 consecutive patients with cBLPDs
treated at our institution with immunochemotherapy in regard to clinical
features and IRR occurence. All patients received standard premedication before
rituximab administration.
Results: IRRs manifested in 24% of all patients. IRR rate was not associated with
gender, age, stage, disease localization (extranodal/nodal), “B” symptoms, bulky
mass, performance status, type of chemotherapy adjunct to rituximab, or rituximab
timing (from 1st cycle vs. from 2nd + cycle). Patients with DLBCL exhibited
significantly fewer IRRs compared to others (p = 0.033), CLL in the first place (15%
vs. 34%, respectively; p = 0.045). Patients treated in 2nd or later therapeutic lines
more frequently exhibited IRR, compared to patients treated in 1st treatment line
(p = 0.0233). Incidence of IRR in patients receiving different rituximab biosimilars
was similar. Arterial hypertension, diabetes mellitus, coronary artery disease,
and smoking did not influence IRR occurrence.
Conclusion: No difference in IRR rates between studied biosimilars was observed.
DLBCL patients and patients receiving rituximab in 1st treatment line suffered
from IRRs less frequently than patients with other cBLPDs and those treated
in 2nd+ line, respectively.Book of abstract: NACIONALNI KONGRES 6. KONGRES HEMATOLOGA SRBIJE, NATIONAL CONGRESS 6TH CONGRESS OF HEMATOLOGISTS OF SERBIA, Hotel Crowne Plaza, Beograd 17–19. oktobar 2024. godin
(PB3059) INFUSION-RELATED REACTIONS IN PATIENTS WITH CHRONIC LYMPHOPROLIFERATIVE DISORDERS TREATED WITH TWO RITUXIMAB BIOSIMILARS: SINGLE CENTER EXPERIENCE
No full textBackground:
One of the most common adverse events of rituximab are infusion-related reactions (IRRs), occurring
in up to 71% of patients (pts), mostly during first infusion. Few percent of patients need to interrupt
further treatment with rituximab due to serious IRR, while 0.07% pts exhibit fatal outcome. Majority
of the data on IRR related to rituximab were provided from clinical trials with originator (Rituxan,
Genetech, Inc./MabThera, Roche), while studies with rituximab biosimilars referring to IRRs are
infrequent. Moreover, there are no published comparisons between different rituximab biosimilars in
context of IRRs.
Aims:
To assess frequency and severity of IRRs to rituximab biosimilars Rixathon® and Blitzima®
administered to patients with chronic lymphoproliferative disorders (LPDs) treated at Clinic for
Hematology, University Clinical Center of Serbia, Belgrade, Serbia from 2020 to 2022.
Methods:
We assessed medical records of 200 consecutive patients with chronic lymphoproliferative disorders
treated at our institution with immunochemotherapy. We examined if there is association between
IRR occurrence and various clinical, radiographic and therapeutic features and compared IRR
frequency between different entities within LPDs, as well as between 2 different rituximab
biosimilars. All pts received premedication consisted of methylprednisolone 1mg/kg, paracetamol 1g,
and loratadine 10mg.
Results:
All demographic and clinical characteristics of patients in our cohort are shown in Table 1.1.
Information about therapeutic modalities and IRR occurrence and manifestations are given in Table
1.2.
Occurrence of IRRs appeared not to be dependent on gender, age, clinical stage, disease localization
(extranodal vs. nodal), presence of “B” symptoms, bulky mass, ECOG performance status, type of
chemotherapy adjunct to rituximab, or rituximab timing (from 1st cycle vs. from 2nd + cycle).
Patients with DLBCL exhibited significantly more often IRR compared to others (P=0.033, χ2 test), CLL
in the first place (P=0.045). Similarly, pts treated in 2nd or later therapeutic lines more frequently
exhibited IRR, compared to pts treated in 1st treatment line (P=0.0233, Fisher exact test). Incidence of
IRR was not significantly different between pts receiving different rituximab biosimilars. Arterial
hypertension (AH), diabetes mellitus (DM), coronary artery disease (CAD), and smoking did not
influence IRR occurrence.
When focusing on subgroup of pts with DLBCL (n=67), we found that pts with AH experienced IRR
more often (P = 0.014, Fisher exact test). Looking closely to pts with CLL (n=76), we noticed more
EHA2024 | Abstract Submission
frequent IRR occurrence in those suffering from AH and DM (P=0.005 and P=0.041, respectively,
Fisher exact test).
Summary/Conclusion:
Incidence of IRR in our cohort was comparable with literature data. DLBCL patients suffered from IRRs
more often than pts with other LPDs. Also, pts receiving rituximab in 2nd+ treatment line more
frequently exhibited IRR. There was no difference in IRR rates between two rituximab biosimilars. In
few patients rituximab was permanently interrupted due to high-grade IRR. Both DLBCL and CLL pts
with AH, and CLL pts also with DM, were at significantly higher risk to develop IRR. IRRs may
complicate treatment with rituximab increasing morbidity and causing treatment interruption in few
percent of patients. Thus, recognizing patients at risk of IRR is an important goal for future studies.29th Congress of the
European Hematology Association
EHA2024 Annual Congress Edition
June 202
Double layer bacterial nanocellulose - poly(hydroxyoctanoate) film activated by prodigiosin as sustainable, transparent, UV-blocking material
No full textThe data and files contained in this dataset are related to the study referring the development of
double-layer UV-blocking film using bacterial nanocellulose (BNC) and poly(hydroxyoctanoic acid) (PHO)
activated with the bacterial pigment prodigiosin (PG), all produced via microbial fermentation. The novel
materials presented a sustainable alternative to synthetic polymers, crucial for reducing plastic waste
and pollution, and it appeared as sustainable, transparent, UV-blocking material suitable for food
packaging application. PHO films with varying PG concentrations (0.2, 0.5, 1.0 wt%) were preferentially
synthesized and characterized to optimize their bioactivity and UV protection. The best-performing
formulation, with low toxicity (tested on MRC5 cells) but preserved desirable UV-blocking ability, was
selected as the outer layer in the preparation of double-layer BNC-PHOPG formulation. The resulting
hydrophilic–hydrophobic bilayer material showed the improved mechanical properties over native BNC.
PG migration from the tested biopolymer films occurred in the most media, except in 10% ethanol.
Overall, all the results highlighted this activated, functional, UV-blocking material as a highly promising
option for use in the packaging industry.readme.txt (5.314Kb)***Dataset contents*** FTIR BNC.csv (134.3Kb)
FTIR BNC-PHOPG2 red side.csv (133.2Kb)
FTIR BN-PHOPG2 white side.csv (131.4Kb)
FTIR BP.csv (131.5Kb)
FTIR PG.csv (34.80Kb)
FTIR PHO.csv (131.5Kb)
FTIR PHOPG and BNC-PHOPG.png (180.5Kb)
FTIR PHOPG1.csv (131.4Kb)
FTIR PHOPG2.csv (131.5Kb)
FTIR PHOPG3.csv (131.4Kb)File readme.txt (5.314Kb) is under licence public domain CC0Dataset for: Malagurski, I., Lazić, J., Ilić-Tomić, T., Salevic, A., Guzik, M., Krzan, M., Nikodinović-Runić, J.,& Ponjavić, M.. (2024). Double layer bacterial nanocellulose - poly(hydroxyoctanoate) film activated by prodigiosin as sustainable, transparent, UV-blocking material. in International Journal of Biological Macromolecules, 279, 135087.
[https://doi.org/10.1016/j.ijbiomac.2024.135087
Evaluation of novel compounds as anti-bacterial or anti-virulence agents
Get PDFAntimicrobial resistance is a global threat, leading to an alarming increase in the prevalence of bacterial infections that can no longer be treated with available antibiotics. The World Health Organization estimates that by 2050 up to 10 million deaths per year could be associated with antimicrobial resistance, which would equal the annual number of cancer deaths worldwide. To overcome this emerging crisis, novel anti-bacterial compounds are urgently needed. There are two possible approaches in the fight against bacterial infections: a) targeting structures within bacterial cells, similar to existing antibiotics; and/or b) targeting virulence factors rather than bacterial growth. Here, for the first time, we provide a comprehensive overview of the key steps in the evaluation of potential new anti-bacterial and/or anti-virulence compounds. The methods described in this review include: a) in silico methods for the evaluation of novel compounds; b) anti-bacterial assays (MIC, MBC, Time-kill); b) anti-virulence assays (anti-biofilm, anti-quorum sensing, anti-adhesion); and c) evaluation of safety aspects (cytotoxicity assay and Ames test). Overall, we provide a detailed description of the methods that are an essential tool for chemists, computational chemists, microbiologists, and toxicologists in the evaluation of potential novel antimicrobial compounds. These methods are cost-effective and have high predictive value. They are widely used in preclinical studies to identify new molecular candidates, for further investigation in animal and human trials