Abu Sayed Chowdhury

Dr. Abu Sayed Chowdhury


Research Focus & Metrics

Academic Career Length

15Years (2012-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 July 2020 in Scientific Reports
Verified
Journal Article
Open access
PDF file
ACAbu Sayed Chowdhury
Abu Sayed Chowdhury
DCDouglas R. Call
Douglas R. Call
SBShira L. Broschat
Shira L. Broschat
With the ever-increasing availability of whole-genome sequences, machine-learning approaches can be used as an alternative to traditional alignment-based methods for identifying new antimicrobial-resistance genes. Such approaches are especially helpful when pathogens cannot be cultured in the lab. In previous work, we proposed a game-theory-based feature evaluation algorithm. When using the protein characteristics identified by this algorithm, called ‘features’ in machine learning, our model accurately identified antimicrobial resistance (AMR) genes in Gram-negative bacteria. Here we extend our study to Gram-positive bacteria showing that coupling game-theory-identified features with machine learning achieved classification accuracies between 87% and 90% for genes encoding resistance to the antibiotics bacitracin and vancomycin. Importantly, we present a standalone software tool that implements the game-theory algorithm and machine-learning model used in these studies.

9 October 2019 in Scientific Reports
Verified
Journal Article
Open access
PDF file
ACAbu Sayed Chowdhury
Abu Sayed Chowdhury
DCDouglas R. Call
Douglas R. Call
SBShira L. Broschat
Shira L. Broschat
The increasing prevalence of antimicrobial-resistant bacteria drives the need for advanced methods to identify antimicrobial-resistance (AMR) genes in bacterial pathogens. With the availability of whole genome sequences, best-hit methods can be used to identify AMR genes by differentiating unknown sequences with known AMR sequences in existing online repositories. Nevertheless, these methods may not perform well when identifying resistance genes with sequences having low sequence identity with known sequences. We present a machine learning approach that uses protein sequences, with sequence identity ranging between 10% and 90%, as an alternative to conventional DNA sequence alignment-based approaches to identify putative AMR genes in Gram-negative bacteria. By using game theory to choose which protein characteristics to use in our machine learning model, we can predict AMR protein sequences for Gram-negative bacteria with an accuracy ranging from 93% to 99%. In order to obtain similar classification results, identity thresholds as low as 53% were required when using BLASTp.

14 August 2023 in Transplantation and Cellular Therapy
Verified
Journal Article
Open access
ACAbu Sayed Chowdhury
Abu Sayed Chowdhury
MMMartin Maiers
Martin Maiers
SSStephen R. Spellman
Stephen R. Spellman
TDTushar Deshpande
Tushar Deshpande
YBYung-Tsi Bolon
Yung-Tsi Bolon
SDSteven Michael Devine
Steven Michael Devine
When patients do not have matched siblings or well-matched unrelated donors (MUD), hematopoietic cell transplantation (HCT) can still be successful using an HLA-mismatched unrelated donor (MMUD), in combination with post-transplant cyclophosphamide (PTCy), abatacept, or other novel approaches. This may allow clinicians to choose a suitable donor from a wide range of donor options while optimizing other donor selection characteristics, including donor age. We hypothesized that allowing for 5/8 HLA match level considering high resolution matching at HLA-A, -B, -C and -DRB1, there is potential to close the donor availability gap for all patients regardless of their race/ethnicity. In this work, we estimate the likelihood of matching for all race/ethnic groups at different HLA match thresholds. Our study aimed to assess the potential for identifying an available MUD or MMUD in the National Marrow Donor Program® (NMDP)/Be The Match® (BTM) donor registry for 21 detailed and 5 broad racial/ethnic groups, utilizing high-resolution HLA matching for HLA-A, B, C, and DRB1 at various levels (8/8, 7/8, 6/8, and 5/8). We used donor registry population data from the NMDP/BTM in 2020 and redistributed the donor registry data according to existing population ratios, accounting for demonstrated donor availability. Finally, we utilized a genetic model at the population level to estimate the match likelihood for detailed and broad racial/ethnic groups. 8/8 HLA match likelihoods ranging from 16-74% were obtained for various detailed race/ethnic groups with available donors ≤35 years old. When considering more mismatches within the HLA loci, registry coverage became >99% with a 5/8 HLA match level for donors of all ages or those ≤ 35 years old, with DPB1 T cell epitope permissive matching, or when searching donors outside of their race/ethnic group. Our registry models demonstrate the potential of using MMUDs at various HLA match levels to study whether this will expand access to HCT across race/ethnic groups. Furthermore, expanded donor options may erase the donor availability gap for all patients while allowing for selection of MMUDs with favorable characteristics such as younger age.

25 October 2019 in Journal of Applied Microbiology
Verified
Journal Article
Open access
PDF file
ACAbu Sayed Chowdhury
Abu Sayed Chowdhury
ELEric T. Lofgren
Eric T. Lofgren
RMRebekah W. Moehring
Rebekah W. Moehring
SBShira L. Broschat
Shira L. Broschat
Aims Analysis and tracking of antimicrobial utilization are crucial in antimicrobial stewardship efforts which are used to find effective interventions for controlling antimicrobial resistance. In antimicrobial stewardship, standard risk adjustment models are needed for benchmarking appropriate antimicrobial utilization and for fair inter‐facility comparison. In this study we identify patient‐ and facility‐level predictors of antimicrobial usage in hospitalized patients using. a machine learning approach, which can be used to inform a risk adjustment model to facilitate assessment of antimicrobial utilization. To our knowledge, this is the first time machine learning. has been applied for this purpose. Methods and Results Patient admission records were retrieved from the Duke Antimicrobial Stewardship Outreach Network which include clinical data for 27 community hospitals in the southeastern United States. Candidate features (predictors) were then generated from these records. The number of features was reduced using a statistical approach, and missing values of the reduced feature set were imputed using bootstrapping and expectation maximization algorithm. Finally, support vector regression (SVR) and cubist regression (CB) models were applied to find root‐mean‐square error values which were used to evaluate the selected feature set. The performance of the SVR and CB models was found to be better than that of linear null and negative binomial null models, thereby demonstrating the effectiveness of our selected features. Conclusions Relevant patient‐ and facility‐level predictors of antimicrobial usage in days of therapy were obtained and evaluated. The potential predictor set can be used in risk adjustment. strategies for benchmarking antimicrobial use. Significance and Impact of the Study One reason for the rapid emergence of antimicrobial resistance is inappropriate use of antibiotics in hospitalized patients. Identifying predictors of antimicrobial exposure using a machine learning technique can improve the use of antimicrobial utilization, enhance patient health outcomes, and reduce the infection spread caused by antimicrobial‐resistant organisms.

17 June 2019 in Journal of Applied Microbiology
Verified
Journal Article
Open access
PDF file
ACAbu Sayed Chowdhury
Abu Sayed Chowdhury
EKE. Khaledian
E. Khaledian
SBS.L. Broschat
S.L. Broschat
Aims Predicting bacterial resistance provides valuable information that can assist in clinical decisions. With recent advances in whole genome sequencing technology, the detection of antibiotic resistance (AR) proteins directly from genomic data is becoming feasible. AR genes/proteins can be identified using best-hit methods that work by comparing candidate sequences with known AR genes in public databases. However, these approaches may fail to detect resistance genes with sequences that differ significantly from known sequences. Our goal is to develop a machine learning technique to accurately predict capreomycin resistance in Mycobacteria with low false discovery rates. Methods and Results We present a stacked ensemble learning model as an alternative to traditional DNA sequence alignment-based methods using optimal features generated from the physicochemical, evolutionary and secondary structure properties of protein sequences. We train logistic regression, C5.0 and support vector machine (SVM) algorithms as our base classifiers, and our stacked ensemble predictors combine the results from the base classifiers to achieve higher accuracy. Compared with our most accurate base classifier (SVM), our most accurate stacked ensemble predictor increases training accuracy by 2·43%. Our stacked ensemble predictors achieve test accuracy up to 81·25%. Conclusions We developed a stacked ensemble model to predict capreomycin resistance for Mycobacteria with an accuracy >80% using protein sequences with sequence similarity ranging between 10% and 70%. This performance cannot be achieved with best-hit methods due to differences in sequence similarity. Significance and Impact of the Study Today an estimated one-half million cases of multidrug-resistant (MDR) and extensively drug-resistant (XDR) tuberculosis (TB) occur annually worldwide at a great cost. Because capreomycin is a second-line drug used to treat drug-resistant TB, the ability to use a machine learning approach to classify capreomycin-resistant TB in a timely manner is crucial for the successful treatment of MDR or XDR TB.

1 August 2015
Verified
Conference Paper
Closed access
ACAbu Sayed Chowdhury
Abu Sayed Chowdhury
MAMonjur Alam
Monjur Alam
YZYanqing Zhang
Yanqing Zhang
It is a great challenge in human health clinic to find parsimonious set of genes responsible for depression disease. Many prioritization approaches have been developed for depression candidate genes. However, most of the methods primarily rank depression candidate genes based on the similarities with known depression genes. Those approaches do not effectively consider relativeness of depression candidate genes with non-disease genes for precise ranking. In this paper, we propose a Biomarker Ensemble Ranking Framework (BERF) for depression candidate gene prioritization, which applies 2-ranking models scheme by considering both candidate genes and non-disease genes. We first clusterize training genes which consists of both known depression and non-disease genes. Then, we introduce a global loss function in the 2-ranking learning model. Finally, we propose a modified SVM-based learning strategy with minimizing the global loss. An ensemble technique is applied to generate the ranking results for depression candidate genes. The experimental results show that our BERF outperforms existing approach in terms of ROC and AUC.