fanyi: Translate Words or Sentences via Online Translators


Useful functions to translate text for multiple languages using online translators. For example, by translating error messages and descriptive analysis results into a language familiar to the user, it enables a better understanding of the information, thereby reducing the barriers caused by language. It offers several helper functions to query gene information to help interpretation of interested genes (e.g., marker genes, differential expression genes), and provides utilities to translate ggplot graphics.

To cite fanyi in publications use:

G Yu. Using fanyi to assist research communities in retrieving and interpreting information. bioRxiv, 2023. doi: 10.1101/2023.12.21.572729.

:writing_hand: Authors

Guangchuang YU

School of Basic Medical Sciences, Southern Medical University

:arrow_double_down: Installation

Get the released version from CRAN:


Or the development version from github:

## install.packages("yulab.utils")

:beginner: Usages

to switch from different online translators:

You should use set_translate_source() to set the default translator using in translate().

Alternatively, you can call specific functions to use a specific translator, (e.g., baidu_translate(), bing_translate(), youdao_translate(), etc.).

to use baidu translate:

  1. go to and regist as an individual developer
  2. get appid and key (密钥)
  3. set appid and key with source = "baidu" using set_translate_option()
  4. have fun with baidu_translate()

to use bing translate:

  1. regist a free azure account
  2. enable Azure AI services | Translator from
  3. create a translation service with free tier pricing version (you need a visa/master card to complete registion and will not be charged until exceed 2 million characters monthly)
  4. get your key and region
  5. set key and region with source = "bing" using set_translate_option()
  6. have fun with bing_translate()

to use youdao(有道智云) translate:

  1. go to and register an account
  2. click 自然语言翻译服务 and create an app from subsection 文本翻译
  3. get your 应用ID as appid, and 应用秘钥 as key
  4. set appid and key with source = "youdao" using set_translate_option()
  5. have fun with youdao_translate()
  6. (bonus) you can also create 术语表 (dictionary for the terms) as a user-defined dictionary and get the dict id to help you get precise translation in certain domain.

to use volcengine (火山引擎) translate:

  1. go to and register an account
  2. enable Machine Translation (机器翻译)
  3. get app key
  4. for security concerns, you are highly advised to add a sub-account (新建子用户)
  5. click the subaccount name in 身份管理 - 用户, and click permissions (权限)
  6. in Global permissions (全局权限), add the following permissions: “TranslateFullAccess”、“I18nTranslateFullAccess”
  7. set appid and key with source = "volcengine" and have fun with volcengine_translate()

to use caiyun (彩云小译) translate:

  1. go to and register
  2. login with your account, create a translation app (创建应用) choose 翻译 as 接口类型
  3. get the token, set it as the key with source = "caiyun" using set_translate_option()
  4. have fun with caiyun_translate()
  5. caiyun currently supports from = {zh(中文), en, ja(日语), auto(自动识别)} and to = {en, ja, zh}, default from is auto

to use tencent (腾讯机器翻译) translate:

  1. go to and register;
  2. login with your account, go to to enable machine translation (启用机器翻译);
  3. go to, create a sub-account, in Permissions (权限) - Permissions policy (权限策略), search and add QcloudTMTReadOnlyAccess;
  4. get your API keys at API keys (API 密钥) of the same page in step 3. set SecretId as the appid and SecretKey as secrets respectively with source = "tencent" using set_translate_option()
  5. have fun with tencent_translate()
  6. for supported languages, visit for more information.

to use chatglm (智谱清言ChatGLM) translate:

  1. go to and register;
  2. login with your account, go to to get your API key;
  3. set the one-line API key as key (no need for setting appid) with source = "chatglm" using set_translate_option()
  4. have fun with chatglm_translate()
  5. you can also set model you prefer to use with the option user_model when using set_translate_option(), currently user_model can be set as glm-4 (by default), glm-4v, glm-3-turbo.

helper functions:

:ideograph_advantage: Examples


## run `set_translate_option()` to setup

text <- '我都不知道做人该怎么办,总之报纸写啥就信啥,电视演啥我就看啥。'
translate(text, from='zh', to='en')

I don’t even know what to do as a person. In short, I believe whatever is written in the newspaper and watch whatever is shown on TV.

translate(text, from='zh', to='th')

ผมไม่รู้ว่าการเป็นมนุษย์ควรทําอย่างไร สรุปแล้วหนังสือพิมพ์เขียนอะไรก็เชื่ออย่างนั้น ทีวีเล่นอะไรก็อ่านอย่างนั้น

translate(text, from='zh', to='jp')


translate(text, from='zh', to='fra')

Je ne sais pas quoi faire en tant que personne, je crois ce que les journaux écrivent, je regarde ce que la télévision fait.

de <- names(geneList)[1:200]
x <- enrichDO(de)
p <- dotplot(x)
p2 <- translate_ggplot(p, axis='y')
p3 <- translate_ggplot(p, axis='y', to='kor')
p4 <- translate_ggplot(p, axis='y', to='ara')
aplot::plot_list(English = p, Chinese = p2, 
                Korean = p3, Arabic = p4, ncol=2)

##  UK: [sel]   US: [sel]
##  Explains: n. 细胞;小牢房;电解槽,电池;电芯;(政治组织的)小组,基层组织;(修道院中的)单人小室;(巢穴中单个的)巢室;(计算机屏幕上的)单元格
##  Web:
symbol <- c("CCR7", "CD3E")
gene <- clusterProfiler::bitr(symbol, 
            fromType = 'SYMBOL', 
            toType = 'ENTREZID', 
            OrgDb = '')

## 1   CCR7     1236
## 2   CD3E      916
res <- gene_summary(gene$ENTREZID)
## [1] "uid"         "name"        "description" "summary"
d <- data.frame(desc=res$description,
##                                             desc                     desc2
## 1                 C-C motif chemokine receptor 7      C-C基序趋化因子受体7
## 2 CD3 epsilon subunit of T-cell receptor complex T细胞受体复合体的CD3ε亚基

[1] The protein encoded by this gene is a member of the G protein-coupled receptor family. This receptor was identified as a gene induced by the Epstein-Barr virus (EBV), and is thought to be a mediator of EBV effects on B lymphocytes. This receptor is expressed in various lymphoid tissues and activates B and T lymphocytes. It has been shown to control the migration of memory T cells to inflamed tissues, as well as stimulate dendritic cell maturation. The chemokine (C-C motif) ligand 19 (CCL19/ECL) has been reported to be a specific ligand of this receptor. Signals mediated by this receptor regulate T cell homeostasis in lymph nodes, and may also function in the activation and polarization of T cells, and in chronic inflammation pathogenesis. Alternative splicing of this gene results in multiple transcript variants. [provided by RefSeq, Sep 2014]

[2] The protein encoded by this gene is the CD3-epsilon polypeptide, which together with CD3-gamma, -delta and -zeta, and the T-cell receptor alpha/beta and gamma/delta heterodimers, forms the T-cell receptor-CD3 complex. This complex plays an important role in coupling antigen recognition to several intracellular signal-transduction pathways. The genes encoding the epsilon, gamma and delta polypeptides are located in the same cluster on chromosome 11. The epsilon polypeptide plays an essential role in T-cell development. Defects in this gene cause immunodeficiency. This gene has also been linked to a susceptibility to type I diabetes in women. [provided by RefSeq, Jul 2008]


[1] 该基因编码的蛋白质是G蛋白偶联受体家族的成员。该受体被鉴定为EB病毒(EBV)诱导的基因,被认为是EB病毒对B淋巴细胞影响的媒介。这种受体在各种淋巴组织中表达,并激活B和T淋巴细胞。它已被证明可以控制记忆T细胞向炎症组织的迁移,并刺激树突细胞成熟。据报道,趋化因子(C-C基序)配体19(CCL19/ECL)是该受体的特异性配体。该受体介导的信号调节淋巴结中的T细胞稳态,也可能在T细胞的激活和极化以及慢性炎症发病机制中发挥作用。该基因的选择性剪接导致多种转录物变体。【RefSeq提供,2014年9月】

[2] 该基因编码的蛋白质是CD3ε多肽,其与CD3γ、-Δ和-ζ以及T细胞受体α/β和γ/Δ异二聚体一起形成T细胞受体-CD3复合物。这种复合物在将抗原识别与几种细胞内信号转导途径偶联方面起着重要作用。编码ε、γ和δ多肽的基因位于11号染色体上的同一簇中。ε多肽在T细胞发育中起着重要作用。这种基因的缺陷会导致免疫缺陷。该基因也与女性易患I型糖尿病有关。【RefSeq提供,2008年7月】

:book: Documents