Biology subjects

Islam, S. M. A.

Publications and source records attributed to Islam, S. M. A..

6 recordsLinked to original sources

bioRxiv2023

Assigning mutational signatures to individual samples and individual somatic mutations with SigProfilerAssignment

Analysis of mutational signatures is a powerful approach for understanding the mutagenic processes that have shaped the evolution of a cancer genome. Here we present SigProfilerAssignment, a desktop and an online computational framework for assigning all types of mutational signatures to individual samples. SigProfilerAssignment is the first tool that allows both analysis of copy-number signatures and probabilistic assignment of signatures to individual somatic mutations. As its computational engine, the tool uses a custom implementation of the forward stagewise algorithm for sparse regression and nonnegative least squares for numerical optimization. Analysis of 2,700 synthetic cancer genomes with and without noise demonstrates that SigProfilerAssignment outperforms four commonly used approaches for assigning mutational signatures. SigProfilerAssignment is freely available at https://github.com/AlexandrovLab/SigProfilerAssignment with a web implementation at https://cancer.sanger.ac.uk/signatures/assignment/.

bioinformatics↗

bioRxiv2021

Comprehensive analysis of mutational signatures in pediatric cancers

Analysis of mutational signatures can reveal the underlying molecular mechanisms of the processes that have imprinted the somatic mutations found in a cancer genome. Here, we present a pan-cancer mutational signatures analysis of single base substitutions (SBS) and small insertion and deletions (ID) in pediatric cancers encompassing 537 whole genome sequenced tumors from 20 molecularly defined cancer subtypes. We identified only a small number of mutational signatures active in pediatric cancers when compared to the previously analyzed adult cancers. Further, we report a significant difference in the proportion of pediatric tumors which show homologous recombination repair defect signature SBS3 compared to prior analyses. Correlating genomic alterations with signature activities, we identified an association of TP53 mutation status with substitution signatures SBS2, SBS8, SBS13 and indel signatures ID2 and ID9, as well as chromothripsis associated with SBS8, SBS40 and ID9. This analysis provides a systematic overview of COSMIC v.3 SBS and ID mutational signatures active across pediatric cancers, which is highly relevant for understanding tumor biology as well as enabling future research in defining biomarkers of treatment response.

cancer biology↗

bioRxiv2020

Uncovering novel mutational signatures by de novo extraction with SigProfilerExtractor

Mutational signature analysis is commonly performed in genomic studies surveying cancer and normal somatic tissues. Here we present SigProfilerExtractor, an automated tool for accurate de novo extraction of mutational signatures for all types of somatic mutations. Benchmarking with a total of 34 distinct scenarios encompassing 2,500 simulated signatures operative in more than 60,000 unique synthetic genomes and 20,000 synthetic exomes demonstrates that SigProfilerExtractor outperforms thirteen other tools across all datasets with and without noise. For genome simulations with 5% noise, reflecting high-quality genomic datasets, SigProfilerExtractor outperforms other approaches by elucidating between 20% and 50% more true positive signatures while yielding more than 5-fold less false positive signatures. Applying SigProfilerExtractor to 4,643 whole-genome sequenced and 19,184 whole-exome sequenced cancers reveals four previously missed mutational signatures. Two of the signatures are confirmed in independent cohorts with one of these signatures associating with tobacco smoking. In summary, this report provides a reference tool for analysis of mutational signatures, a comprehensive benchmarking of bioinformatics tools for extracting mutational signatures, and several novel mutational signatures including a signature putatively attributed to direct tobacco smoking mutagenesis in bladder cancer and in normal bladder epithelium.

bioinformatics↗

bioRxiv2019

Making plants into cost-effective bioreactors for highly active antimicrobial peptides

As antibiotic-resistant bacterial pathogens become an ever-increasing concern, antimicrobial peptides (AMPs) have grown increasingly attractive as alternatives. Potentially, plants could be used as cost-effective AMP bioreactors; however, reported heterologous AMP expression is much lower in plants compared to E. coli expression systems and often results in plant cytotoxicity, even for AMPs fused to carrier proteins. We wondered if there were a physical factor that made heterologous AMPs difficult to express in plants. Using a meta-analysis of protein databases, we determined that native plant AMPs were significantly less cationic than AMPs native to other taxa. To apply this finding to plant expression, we tested the transient expression of 10 different heterologous AMPs, ranging in charge from +7 to -5, in the the tobacco, Nicotiana benthamiana. We first tested several carrier proteins and were able to express AMPs only with elastin-like polypeptide (ELP). Conveniently, ELP fusion allows for a simple, cost-effective temperature shift purification. Using the ELP system, all five anionic AMPs expressed well, with two at unusually high levels (375 and 563 {micro}g/gfw). Furthermore, antimicrobial activity against Staphylococcus epidermidis was an order of magnitude stronger (average MIC = 0.26 {micro}M) than that typically seen for AMPs expressed in E. coli expression systems. Unexpectedly, this high level of antimicrobial activity was associated with the uncleaved fusion peptide. In contrast, all previous reports of AMPs expressed in both plant and E. coli expression systems show cleavage from the fusion partner to be required before activity is seen. In summary, we describe a means of expressing AMP fusions in plants in high yield, purified with a simple temperature-shift protocol, resulting in a fusion peptide with high antimicrobial activity, without the need for a peptide cleavage step.

plant biology↗

bioRxiv2019

The use of a virus-derived targeting peptide to selectively kill staphylococcus bacteria with antimicrobial peptides

BackgroundTargeted therapies seek to selectively eliminate a pathogen without disrupting the microbiome community. Bacteriophages provide a rich, well-documented source of bacterium-specific binding proteins for use as targeting peptides fused to antimicrobial peptides. Though resistance may develop as with any antibiotic, the wealth of variants available in natural bacteriophage populations adds to the robustness of this system.\n\nResultsHere, we target two cationic antimicrobial peptides (AMPs), plectasin and eurocin, by genetically fusing their coding sequence to that of the host-binding protein of bacteriophage A12C, which selectively infects Staphylococcus. Surprisingly, we noted that targeting brought no change in the toxicity of the AMP when applied to two different staphylococci, S. aureus and S. epidermidis, but found a drastic decrease in toxicity against the negative controls, Enterococcus faecalis and Bacillus subtilis. Thus, the differential selectivity in this case is a loss of toxicity against the non-target species rather than the gain of toxicity against the target species which was reported in previous studies with other types of targeting antimicrobial peptides.\n\nConclusionThis is the first report of the use of virus-derived peptide sequences to target antimicrobial peptides. Considering the very large databank of bacteriophages and their bacterial hosts, this targeting approach should be generally applicable to a wide range of bacterial pathogens.

molecular biology↗

bioRxiv2017

Protein classification using modified n-gram and skip-gram models

MotivationClassification by supervised machine learning greatly facilitates the annotation of protein characteristics from their primary sequence. However, the feature generation step in this process requires detailed knowledge of attributes used to classify the proteins. Lack of this knowledge risks the selection of irrelevant features, resulting in a faulty model. In this study, we introduce a means of automating the work-intensive feature generation step via a Natural Language Processing (NLP)-dependent model, using a modified combination of N-Gram and Skip-Gram models (m-NGSG).\n\nResultsA meta-comparison of cross validation accuracy with twelve training datasets from nine different published studies demonstrates a consistent increase in accuracy of m-NGSG when compared to contemporary classification and feature generation models. We expect this model to accelerate the classification of proteins from primary sequence data and increase the accessibility of protein prediction to a broader range of scientists.\n\nAvailabilitym-NGSG is freely available at Bitbucket: https://bitbucket.org/smislam/mngsg/src\n\nSupplementslink to supplementary documents\n\nContactErich_Baker@baylor.edu

bioinformatics↗