Search

Having a central scientific language remains crucial for advancing and globally sharing science. Nevertheless, maintaining one dominant language also creates barriers to accessing scientific careers and knowledge. From an interdisciplinary perspective, we describe how, when, and why to make scientific literature more readily available in multiple languages through the practice of translation. We broadly review the advantages and limitations of neural machine translation systems and propose that translation can serve as both a short- and a long-term solution for making science more resilient, accessible, globally representative, and impactful beyond the academy. We outline actions that individuals and institutions can take to support multilingual science and scientists, including structural changes that encourage and value translating scientific literature. In the long term, improvements to machine translation technologies and collective efforts to change academic norms can transform a monolingual scientific hub into a multilingual scientific network. Translations are available in the supplemental material.

During the last 20 years, our group has focused on identifying the genes and mutations causative of inherited retinal dystrophies (IRDs). By applying massive sequencing approaches (NGS) in more than 500 familial and sporadic cases, we attained high diagnostic efficiency (85%) with a custom target gene panel and over 75% using whole exome sequencing (WES). Close to 40% of pathogenic alleles are novel mutations, which demand specific in silico tests and in vitro assays. Notably, missense variants are by far the most common type of mutation identified (around 40%), with small in-frame indels being less frequent (2%). To fill the gap of unsolved cases, when no candidate gene or only a single pathogenic allele has been identified, additional scientific and technical issues remain to be addressed. Reliable detection of genomic rearrangements and copy number variants (partial or complete), deep intronic mutations, variants that cause aberrant splicing events in retina-specific transcripts, functional assessment of hypomorphic missense alleles, mutations in regulatory sequences, the contribution of modifier genes to the IRD phenotype, and detection of low heteroplasmy mtDNA mutations are among the new challenges to be met.

Treballs Finals de Màster de Comunicació Especialitzada, Facultat de Filologia, Universitat de Barcelona, Curs: 2018-2019, Tutora: Maria Carme Junyent Figueras, [eng] Currently, the success of a scientist depends on the rate of productivity of articles and the journal impact factor of its research. This implies publishing in English. Currently, more than 90% of science publications are written in English, including among researchers from non-English-speaking countries. Among the peripheral countries, Colombia is the third most unequal territory in the world and invests only 0.67% of its GDP in science and technology. This country is also placed among the countries with the lowest English proficiency in the world. Understanding what are the disadvantages faced by scientists in this country in the production of articles is crucial to reduce global inequality in this field. The main objective of this work is to quantify the disadvantages that English generates on the production of scientific articles and oral presentations in 49 Colombian doctoral students of biological sciences. Among the results, it was determined that publishing in English creates additional costs for doctoral students at an economic level, anxiety level, reading comprehension, ease and writing time, among others. Additionally, there is a relationship between some of the costs and the socioeconomic origin of the researcher. Finally, the experience of the TIPPs service of the Universidad de los Andes, Bogotá, shows us a possibility to face the problem from the maintenance of monolingualism in science, however, in this study we discuss how other alternatives are possible., [spa] En el presente, el éxito de un científico depende de su tasa de productividad de artículos y el factor de impacto de la revista donde se publique, esto implica publicar en inglés. Actualmente, más del 90% de las publicaciones en ciencia son escritas en inglés, inclusive entre los investigadores de países no angloparlantes nativos. Entre los países periféricos, Colombia es el tercer territorio más desigual del mundo e invierte únicamente un 0.67% de su PIB en ciencia y tecnología, hecho que lo sitúa entre los países con más bajo dominio de inglés en el mundo. Entender cuáles son las desventajas que enfrentan los científicos de este país en la producción de artículos es crucial para disminuir la desigualdad global en este campo. El objetivo principal de este trabajo es cuantificar las desventajas que genera el inglés sobre la producción de artículos científicos y presentaciones orales en 49 doctorandos colombianos de ciencias biológicas. Entre los resultados se determinó que publicar en inglés crea costos adicionales sobre los doctorandos a nivel económico, nivel de ansiedad, compresión de lectura, facilidad y tiempo de escritura, entre otros. Adicionalmente, que existe una relación entre algunos de los costos y el origen socioeconómico del investigador. Por último, la experiencia del servicio TIPPs de la Universidad de los Andes, Bogotá, nos muestra una posibilidad para hacer frente al problema desde el mantenimiento del monolingüismo en la ciencia, no obstante, en este estudio se discute como otras alternativas son posibles.

Andinobates victimatus sp. nov. Figures 2 and 4 A–K. Table 1. Holotype. An adult female (ANDES-A 3686) collected by Daniel Mejía-Vargas in Chichiridó, Dabeiba, Antioquia (7.02°, -76.38°), located in the Llorona Canyon, at 280 meters above sea level (m.a.s.l.). Paratypes. Five adults (4 males and 1 female; ANDES-A 3684, 3685, 3687–3689), collected at the type locality, along with the holotype. Etymology. From the Latin victimatus (the victimized); noun in apposition. The Urabá region, where this species occurs, has historically been flailed by Colombia’s longstanding armed conflicts, perhaps more than most other regions of the country, leaving a virtually countless trail of innocent victims. We name this species in honor and remembrance of all these victims. Definition and diagnosis. Andinobates victimatus sp. nov. is a minute (SVL mean = 14.23 ± SD = 0.67 mm; Table 1) dendrobatid frog, characterized by having homogeneous scarlet red dorsal coloration, which extends throughout the limbs. Fingers and toe tips are light metallic gray to metallic gray. The venter and flanks are also scarlet red, with patches degrading into vermilion and finally black, forming irregular blotches (Fig. 4 A–F). The skin is slightly granular, with granularity increasing towards the urostyle and hind limbs. This species is assignable to Andinobates based on the following traits: small adult size (SVL Andinobates, we place A. victimatus sp. nov in the A. fulguritus group (Brown & Twomey et al. 2011), also based on phylogenetic analyses, juvenile color pattern (Fig. 4 H–K; see Coloration in life), and calls of the “short buzz” type (Brown & Twomey et al. 2011; Fig. 5). The new species can be easily distinguished from most other known members of Andinobates by its distinctive solid red dorsal coloration (Fig. 4). The only exception is the recently described A. geminisae (Batista et al. 2014), which has uniform orange dorsal coloration, but lacks light gray/tan pigmentation on finger and toe tips, has uniform ventral coloration, and produces calls of the “tonal buzz” type, with distinctive clicks at the beginning and end of the call, as opposed to the “short buzz” calls of A. victimatus. Moreover, calls of A. victimatus sp. nov. are shorter in duration (0.42– 0.58 s vs. 1.52– 1.6 s) and higher in peak frequency (5.65–5.86 kHz vs. 4.40–4.50 kHz; Fig. 6) than those of A. geminisae. Finally, A. geminisae is, on average, smaller than A. victimatus sp. nov. (SVL 12.68 ± 0.87 mm vs. 14.23 ± 0.67 mm; t 15 = 3.78, p = 0.002, N = 6 A. vic, 11 A. gem; Fig. 6). However, the SVL ranges of both species overlap slightly (A. geminisae: 11.63–13.63 mm, A. victimatus sp. nov.: 13.22–14.98 mm). Some individuals of A. opisthomelas, A. virolinensis, and A. cassidyhornae can have nearly homogenous red dorsal coloration (Amézquita et al. 2013; Ruiz-Carranza & Ramírez-Pinilla 1992; Silverstone 1975). However, they can be distinguished from A. victimatus sp. nov. by their ventral pattern which consists of red blotches on a black background in A. cassidyhornae, light gray spots/marbling in A. virolinensis and some populations of A. opisthomelas, and dark brown with irregular suffusions of red from the flanks in other populations of A. opisthomelas. Moreover, the three species are, on average, larger than A. victimatus (A. cassidyhornae: 19.031 ± 0.31 mm, t 11 = 17.04, p A. vic, 7 A. cas; A. virolinensis: 16.86 ± 0.91 mm, t 58 = 6.85, p A. vic, 54 A. vir; A. opisthomelas: 16.80± 1.24 mm, t 53 = 5.14, p A. vic, 49 A. opi; data from original descriptions and Silverstone 1975), although, again, with slight range overlap with A. virolinensis (13.22– 14.98 mm vs. 14.6–18.9 mm) and A. opisthomelas (13.22–14.98 mm vs. 14.5–19.5 mm). In addition to coloration, A. victimatus sp. nov. can be distinguished from other lowland species of Andinobates (i.e. A. minutus and A. fulguritus groups) in the following ways (Fig. 6): The calls of A. victimatus sp. nov. are shorter than those of A. claudiae (0.42– 0.58 s vs. 0.85– 1.03 s), and have a higher peak frequency than those of A. fulguritus (5.65–5.86 kHz vs. 4.83–5.16 kHz); A. victimatus sp. nov. is smaller in body size than A. altobueyensis (13.22–14.98 mm vs. 16.30–17.00 mm). Description of the holotype. An adult female, 13.22 mm in snout-vent length. Snout dorsally subtruncate, laterally rounded and slightly protrusive; canthus rostralis rounded; loreal region vertical and slightly concave. The head is wider than it is long, and the same width as the body (HW/GBW = 1.002); head width 29.5% of SVL; head length 25.0% of SVL; nares 1.24 times closer to tip of snout than to eyes; pupil rounded and horizontally elliptical; eye length 52.6% of head length. Tympanic ring visible, oval shaped, of moderate size (HDT 41.4% of EL); posterior and dorsal margins hidden under skin. Maxillary and premaxillary teeth absent; median lingual process absent; tongue about two times longer than wide; anterior margin of tongue not indented; posterior third of tongue not attached to floor of mouth. Hands of moderate size (HaL 26.6% of SVL); adpressed finger lengths are I In life, the dorsum was homogeneous scarlet red, with faintly visible dorsolateral stripes; flanks were mostly uniform vermilion, with a few small, dark specks towards the ventral area. The venter was also vermilion with lesssaturated blotches transitioning from vermilion to black towards the chest and throat. Fingers and toes were light metallic gray, and palms and soles were black. The skin was slightly granular, especially in the posterior venter, cloaca, and dorsal surface of hind limbs. In ethanol (70%), the dorsum and dorsal surfaces of limbs are metallic gray; hands, feet, and articulations are tan to light brown. The venter and flanks are also metallic gray, with some scattered dark brown specks; a large dark brown blotch is present on the throat. The skin is mostly smooth, except for the ventral surface of the hind limbs and cloaca, where weak granularity is still visible. Coloration in life (Figs. 4, 4 A–K). The dorsum and dorsal surfaces of limbs of all individuals in the type series are homogenous scarlet red. In some collected and uncollected individuals, a faint, complete dorsolateral stripe (type A in Grant et al. 2006) of a more saturated shade of red is visible (Fig 4 A-D). The scarlet red becomes less saturated in the flanks and venter, transitioning between scarlet-red, vermilion, and black in gradually changing blotches, especially in the throat, chest, and the dorsal surface of the thighs (forming a seat patch; Fig 4 E–F). The pupil and iris are both black, and nearly indistinguishable. The nares are surrounded by thin black margins, and the anterior edge of the tympanic ring is also black, with the first quarter of the tympanum degrading from black to vermilion. Fingers and toes are light metallic gray to metallic gray, and palms and soles black to dark brown. Juveniles reared in the laboratory (Fig. 4 H–K) developed tan/orange dorsolateral stripes as metamorphs (around stage 40), and later acquired a faint incomplete dorsal median stripe that extended from the tip of the snout to the posterior-dorsal region, and a labial stripe extending from just below the snout to the insertion of the arm, both reminiscent of the color pattern of some A. fulguritus. As time went by, the stripes became a more saturated red, and the dark brown background gave way to homogenous scarlet red pigmentation. Adult coloration was attained around 3.5 months after leaving the water. These individuals were reared at a lower temperature (~18–22 °C) than their natural habitat (~22–28 °C), so this process is probably quicker in nature. Coloration in preservative (70% ethanol). In preservative, the red color turns metallic gray. Fingers, toes, and tubercles become light tan/brown color, and the pupil and cornea become light gray to white. The heterogeneity in ventral coloration becomes more defined in preservative, as the black blotches become dark brown, or stay black, while the scarlet red/vermilion background becomes metallic gray. Vocalization (Figs. 5–6). Similar to other members of the A. fulguritus group, A. victimatus sp. nov. produces calls of the “short buzz” type, consisting of a series of pulses modulated in amplitude but not frequency (Brown & Twomey et al. 2011; Myers & Daly 1976). Calls range from 0.42– 0.58 s in length (mean = 0.49 ± SD = 0.06 s), and contain 99–190 pulses per call (142 ± 42.6). The peak frequency ranges between 5.65–5.86 kHz (5.75 ± 0.08 kHz), and the frequency interquartile bandwidth between 0.76–1.03 kHz (0.88 ± 0.11 kHz). Natural history. A. victimatus sp. nov. is a terrestrial species, which inhabits the understory of lowland forests in the Urabá region of Northwestern Colombia (Fig. 7). Males call to defend territories and attract mates, and were typically observed doing so from the leaf litter, or perched on low branches ( Egg clutches are presumably laid within male territories, and tadpoles are transported by males, one at a time, to bromeliad tanks. Tadpoles were observed in bromeliads up to 4m above the ground (although we did not survey plants above this height). All of the observed tadpoles were in tanks of Guzmania musaica bromeliads (Fig. 7 D), which leads us to speculate that A. victimatus sp. nov. prefers this bromeliad over other similarly-sized species (e.g. other Guzmania or Tillandsia), for yet unknown reasons. At the type locality, A. victimatus sp. nov. is sympatric with Oophaga histrionica (Anura: Dendrobatidae), which deposits larvae in similar phytotelmata. However, we did not find any O. histrionica tadpoles in G. musaica tanks; only in smaller bromeliad species. Whether this indicates some instance of resource partitioning, competition, or another kind of ecological interaction between the two species remains to be studied. Distribution and conservation status. This species occurs in the Urabá region of Northwestern Colombia, inhabiting tropical rain forests at ~ 200– 600 m.a.s.l. (Figs. 1, 7) in the foothills of the Western Andes, west of the Paramillo Massif, on the Atrato River drainage. The type locality is in the Llorona Canyon in Dabeiba, Antioquia, Colombia, just southwest of the Massif. We have become aware of reports of similar frogs from other biologists and conservationists working in the Urabá region (Carlos Bran-Castrillón, pers. com.), which would extend the new species’ distribution north into the municipalities of Mutatá and Carepa, almost to the Gulf of Urabá, on the Caribbean coast (Fig. 1). In addition, two specimens at the Natural History Museum of the Universidad de Antioquia (MHUA-A 3382 and 3864), currently assigned to A. opisthomelas, are from the municipality of Tierralta, Córdoba, at ~ 590 m. a.s.l (Fig. 1). It is possible that these two specimens actually correspond to A. victimatus sp. nov., since A. opisthomelas does not normally occur below 1100 m.a.s.l (Silverstone, 1975), which would mean that this species is distributed around the northern end of the Paramillo Massif, into the Sinú River drainage. As we have not examined specimens, alive or preserved, from any of these localities, we consider them tentative reports. Andinobates victimatus sp. nov. has, to the best of our knowledge, not been reported south or west of the type locality. However, given the continuous rain forests and gentle topography of the Northeastern Chocó, it is likely that this species does extend into these areas. Nonetheless, Silverstone did not find it when collecting A. fulguritus in the Arquía river, ~ 85 km south of the type locality, in the 1960s, and it has not been reported in the vicinities of Quibdó (~ 130 km south) or the San Juan River drainage (> 180 km south), both well sampled in terms of Dendrobatid frogs. Therefore, we suspect that this species’ range does not extend too far south of the type locality. Further exploration and inspection of museum specimens are needed to get a better grasp of this species’ geographic distribution. In spite of the current uncertainty regarding the distribution of A. victimatus sp. nov., it seems clear that its extent of occurrence is less than 5,000 km 2. It is currently known from one locality with certainty, plus five tentative additional ones, and the forests in this species’ range have been, and will probably continue to be under strong deforestation pressures due to mining (both legal and illegal), cattle farming, and large-scale agriculture (e.g. banana and plantain plantations), which reduce the amount and quality of available habitat, and increase its fragmentation. Therefore, we tentatively designate A. victimatus sp. nov. as Endangered (EN: B1a, biii; IUCN, 2001). However, it is possible that the range of A. victimatus sp. nov. does exceed 5,000 km 2, in which case this species should be listed as Vulnerable (VU; B1a, biii). Most of the range of A. victimatus sp. nov. surrounds the Nudo de Paramillo National Natural Park, a 460,000 ha nature reserve, managed by the Colombian government, which is closed to the public. However, we are unaware of any reports of A. victimatus sp. nov within the park, which encompasses most of the Paramillo Massif above 700 m. a.s.l. This national park, nonetheless, protects ~250,000 ha in the 700–1000 m.a.s.l elevation belt, where A. victimatus could exist at the higher end of its elevational range. Considerably more research on this species’ distribution, habitat requirements, and population ecology is needed to confidently assign it to a category of threat. Voucher Species Locality Study GenBank Accession 16S COI Cytb ……continued on the next page Voucher Species Locality Study GenBank Accession 16S COI Cytb GECOH 1216 Andinobates opisthomelas I la đel SƟl, Guatapé, AntiƟquia, Amézquita et al. JQ936634 KY885051 JQ936620 CƟlƟmbia 2013 GECOH 297 Andnobates bombetes Bosque YƟtƟcƟ Natural Re erve, YƟtƟcƟ, Thi tuđy KY885045 KY885067 KY885092 Valle đel Cauca, CƟlƟmbia GECOH 305 Andinobates virolinensis CƟ tilla đe fara, VirƟlín, Santanđer, ƂrƟwn et al. 2011; JN635873 KY885068 JQ936626 CƟlƟmbia Amézquita et al. 2013 GECOH 474 Andinobates tolimensis Falan, TƟlima, CƟlƟmbia ƂrƟwn et al. 2011; JN635857 KY885069 JQ936632 Amézquita et al. 2013 GECOH 1523 Andinobates cassidyhornae La Me enia, Jarđín, AntiƟquia, CƟlƟmbia Amézquita et al. JQ936636 KY885054 JQ936622 2013 MVUP 2428 Andinobates geminisae 6 Km ea t Ɵf El Carmen (RíƟ Ƃelén), Ƃati ta et al. 2014 KM212167 KM212166 - Heađwater Ɵf RíƟ CanƟ, DƟnƟ Ɵ, CƟlón, Panamá USNM-FS Andinobates claudiae SƟuth enđ Ɵf I la PƟpa, 1km E Ɵf FrƟ t et al. 2004; DQ283042 DQ502748 DQ502453 59980 SunwƟƟđ Channel, ƂƟca đel TƟrƟ, Grant et al. 2006 Panamá Paratype; ** HƟlƟtype