Large Language Model and Large Vision Model for Life Sciences

Dear Colleagues,

Artificial intelligence (AI) has profoundly changed the research paradigm of life sciences. Many Large Language Models (LLMs) inspired by ChatGPT have been built to accelerate biomedical studies. For example, GeneGPT can answer biomedical questions uses NCBI Web APIs and CancerGPT can predict drug pair synergy using large pre-trained language models.

Beside text data, there are many images and videos in biomedical researches, such as pathology images, light or electron microscopy cell images and videos. It is estimated that over one million digital pathology images are collected per day all over the USA. Many hospitals are sharing their large repositories. Large Vision Model (LVM) can be used to analyze such large volume image and video data.

Since LLM and LVM are trained on large volume data with large diversity, they are general models and can do a lot of tasks without too much sample and feature engineering. For language problems, we can build an NLP (Natural Language Processing) model to do translation and build another NLP model to fix grammar errors. But for LLM like ChatGPT which are trained on almost the entire Internet data, one model can do all these tasks. Similarly, we can train a traditional CV (Computer Vision) model between the culture plate images and the number of cells. But for another cell type with different shapes, we need to train another model. Now, we can train one LVM and give a prompt of a series of cell A growth images and few cell B growth images and ask the LVM model to predict in next image how many B cells will be.

Although it is still challenging to build LVM, several new approaches such as visual sentences have been proposed. The visual data can be represented as sequences, the LVM can be trained by minimizing the cross-entropy loss for next token prediction. Different vision tasks can be solved with suitable visual prompts.

The fusion of LLM and LVM makes the Large Multimodal Model (LMM) possible and eventually the Artificial General Intelligence (AGI) possible. We will understand its underlying meaning no matter we see a sentence or an image or a video.

Therefore, we invite multidisciplinary researchers to submit original articles, as well as review articles on this special issue. Potential topics of interest include, but are not limited to, the following:

1. Large language model for single cell studies
2. Large language model in protein design
3. Large language model in High Throughput Drug screening/ discovery
4. Large vision model for pathology images
5. Large vision model for spatial omics images
6. Large vision model for cancer cell growth and migration
7. Large multimodal model for building digital twins of cancer progression
8. Large vision model in High Throughput Drug screening/ discovery

Dr. Tao Huang
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Life is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• artificial intelligence
• large language model
• large vision model
• images
• videos
• single cell
• protein design
• spatial omics
• drug discovery
• digital twin