Taha ValizadehAslani

Dr. Taha ValizadehAslani

Liang Lab
orcid

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

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

Representative Publications

View All Publications
28 October 2020 in Biology
Verified
Journal Article
Open access
PDF file
Taha ValizadehAslaniTaha ValizadehAslani
Taha ValizadehAslani
ZZZhengqiao Zhao
Zhengqiao Zhao
BSBahrad Sokhansanj
Bahrad Sokhansanj
GRGail L. Rosen
Gail L. Rosen
Machine learning algorithms can learn mechanisms of antimicrobial resistance from the data of DNA sequence without any a priori information. Interpreting a trained machine learning algorithm can be exploited for validating the model and obtaining new information about resistance mechanisms. Different feature extraction methods, such as SNP calling and counting nucleotide k-mers have been proposed for presenting DNA sequences to the model. However, there are trade-offs between interpretability, computational complexity and accuracy for different feature extraction methods. In this study, we have proposed a new feature extraction method, counting amino acid k-mers or oligopeptides, which provides easier model interpretation compared to counting nucleotide k-mers and reaches the same or even better accuracy in comparison with different methods. Additionally, we have trained machine learning algorithms using different feature extraction methods and compared the results in terms of accuracy, model interpretability and computational complexity. We have built a new feature selection pipeline for extraction of important features so that new AMR determinants can be discovered by analyzing these features. This pipeline allows the construction of models that only use a small number of features and can predict resistance accurately.

9 January 2023 in Journal of Biomedical Informatics
Verified
Journal Article
Closed access
YSYiwen Shi
Yiwen Shi
JWJing Wang
Jing Wang
PRPing Ren
Ping Ren
Taha ValizadehAslaniTaha ValizadehAslani
Taha ValizadehAslani
YZYi Zhang
Yi Zhang
Hualou LiangHualou Liang
Hualou Liang

21 October 2020 in Journal of Circadian Rhythms
Verified
Journal Article
Open access
PDF file
CCChiahui Chen
Chiahui Chen
Taha ValizadehAslaniTaha ValizadehAslani
Taha ValizadehAslani
GRGail L. Rosen
Gail L. Rosen
LALaura M. Anderson
Laura M. Anderson
CJCarla R. Jungquist
Carla R. Jungquist
Background: Circadian misalignment can impair healthcare shift workers' physical and mental health, resulting in sleep deprivation, obesity, and chronic disease. This multidisciplinary research team assessed eating patterns and sleep/physical activity of healthcare workers on three different shifts (day, night, and rotating-shift). To date, no study of real-world shift workers' daily eating and sleep has utilized a largely-objective measurement. Method: During this fourteen-day observational study, participants wore two devices (Actiwatch and Bite Technologies counter) to measure physical activity, sleep, light exposure, and eating time. Participants also reported food intake via food diaries on personal mobile devices. Results: In fourteen (5 day-, 5 night-, and 4 rotating-shift) participants, no baseline difference in BMI was observed. Overall, rotating-shift workers consumed fewer calories and had less activity and sleep than day-and night-shift workers. For eating patterns, compared to night-and rotating-shift, day-shift workers ate more frequently during work days. Night workers, however, consumed more calories at work relative to day and rotating workers. For physical activity and sleep, night-shift workers had the highest activity and least sleep on work days. Conclusion: This pilot study utilized primarily objective measurement to examine shift workers' habits outside the laboratory. Although no association between BMI and eating patterns/activity/sleep was observed across groups, a small, homogeneous sample may have influenced this. Overall, shift work was associated with 1) increased calorie intake and higher-fat and-carbohydrate diets and 2) sleep deprivation. A larger, more diverse sample can participate in future studies that objectively measure shift workers' real-world habits.

1 December 2018 in Physical Communication
Verified
Journal Article
Closed access
Taha ValizadehAslaniTaha ValizadehAslani
Taha ValizadehAslani
AFAbolfazl Falahati
Abolfazl Falahati

11 September 2020 in bioRxiv
Verified
Preprint
Open access
PDF file
WIWaleed Iqbal
Waleed Iqbal
EDElena Demidova
Elena Demidova
SSSamantha Serrao
Samantha Serrao
Taha ValizadehAslaniTaha ValizadehAslani
Taha ValizadehAslani
GRGail L. Rosen
Gail L. Rosen
SASanjeevani Arora
Sanjeevani Arora
RRM2B plays a crucial role in DNA replication, repair and oxidative stress. While germline RRM2B mutations have been implicated in mitochondrial disorders, its relevance to cancer has not been established. Here, using TCGA data, we investigated RRM2B alterations in cancer. We found that RRM2B is highly amplified in multiple tumor types, particularly in MYC -amplified tumors, and is associated with increased RRM2B mRNA expression. We also observed that the chromosomal region 8q22.3–8q24, is amplified in multiple tumors, and includes RRM2B , MYC along with several other cancer-associated genes. An analysis of genes within this 8q-amplicon showed that cases that have both RRM2B -amplified along with MYC have a distinct pattern of amplification compared to unaltered cases or cases that have amplifications in RRM2B or MYC only. These other 8q-proteins were shown to interact functionally within the RRM2B network of DNA repair, hypoxia and apoptosis regulating proteins. Notably, RRM2B -amplified tumors are characterized by mutation signatures of defective DNA repair and oxidative stress, and in some cancers also associated with poor clinical outcome. These findings suggest that some cancers may require RRM2B for cellular survival, providing novel therapeutic opportunities in these cancers.

16 April 2024
Verified
Thesis
Closed access
Taha ValizadehAslaniTaha ValizadehAslani
Taha ValizadehAslani
Hualou LiangHualou Liang
Hualou Liang
GRGail L. Rosen
Gail L. Rosen
Large language models (LLMs) have emerged as a milestone in the realm of machine learning, offering groundbreaking capabilities in understanding human language with remarkable coherence and versatility. Their application spans a multitude of domains, ranging from text classification to information extraction. Such models are first pre-trained on vast data corpora to learn general language patterns and then are fine-tuned on specific downstream tasks. Despite their impressive capabilities, LLMs encounter several limitations, including sub-optimal performance in understanding the domain-specific language of drug labels, limited performance in analyzing class-imbalanced data, and high computational complexity of training. Human prescription drug labeling contains a summary of the essential scientific information needed for the safe and effective use of the drug and includes critical information about drug products, such as pharmacokinetics and adverse events. Automatic information extraction from drug labels can facilitate processes such as finding the adverse reaction of the drugs or finding the interaction of one drug with another drug. Although LLMs are an interesting solution to perform this task, because of their training on generic data, they may not possess the depth of knowledge or the ability required to understand the language in such a specialized field. To address this issue, this work presents the developed PharmBERT, which is a BERT model specifically pretrained on the drug labels, publicly available online for the NLP community. We demonstrate that our model outperforms models such as vanilla BERT, ClinicalBERT, and BioBERT in multiple NLP tasks in the drug label domain. Another shortcoming of LLMs lies in their handling of class-imbalanced data, leading to a subpar performance in identifying and accurately responding to minority classes. This class-imbalance can skew the model's learning process, favoring majority classes, and thus diminishing its effectiveness in recognizing and appropriately addressing less represented categories. To cope with these challenges, we present a simple modification of standard fine-tuning. Specifically, we propose a two-stage fine-tuning. In Stage 1, we fine-tune the final layer of the pretrained model with class-balanced augmented data, generated using ChatGPT. As a large generative language model, ChatGPT is capable of generating novel yet contextually similar responses to a given text sample, which makes it an excellent candidate for data augmentation. In Stage 2, we perform the standard fine-tuning. Our modification has several benefits: (1) it leverages pretrained representations by only fine-tuning a small portion of the model parameters while keeping the rest untouched; (2) it allows the model to learn an initial representation of the specific task; and importantly (3) it protects the learning of tail classes from being at a disadvantage during the model updating. The experimental results show that the proposed two-stage fine-tuning outperforms vanilla fine-tuning and state-of-the-art methods on the above datasets. The third issue of LLMs is the high computational complexity of training. Due to the large number of parameters in many state-of-the-art LLMs, such as Bidirectional Encoder Representations from Transformers (BERT), the process of fine-tuning exhibits a high level of intricacy. One attractive solution to this issue is parameter-efficient fine-tuning, which means only fine-tuning a small portion of the model and not changing the rest. However, the question that which portion of the BERT model is the most important one for fine-tuning remains unanswered. Here, we first analyze different components in the BERT model to see which one undergoes the most change after fine-tuning. We show that output LayerNorm changes more than other components when finetuned for different General Language Understanding Evaluation (GLUE) tasks. Then we show that only fine-tuning this component reaches a comparable, or in some cases better, performance to full fine-tuning and other parameter-efficient fine-tuning methods. Moreover, we use Fisher information to find out which parameters in LayerNorm are the most important ones and show that with a small performance degradation, many NLP tasks in the GLUE benchmark can be solved by fine-tuning only a small, selected portion of LayerNorm. In summary, LLMs hold immense potential to redefine and enhance various fields, including drug-label analysis, but they also suffer from limitations like sub-optimal performance in understanding domain-specific texts of drug labels, not learning minority classes in the class-imbalanced data, and high computational complexity. In this thesis, for each one of these shortcomings, first, the problem is analyzed and then a solution is provided.